Guidance Document – Cost-Benefit-Analysis in freshwater ecosystem restoration

Deliverable D3.4: Guidance Document – CostBenefit-Analysis in freshwater ecosystem restoration www.project-merlin.eu MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 2 Imprint The MERLIN project (https://project-merlin.eu) has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 101036337. Lead contractor: Wageningen University To be cited as: Kok S., Grondard N., Lenz M.I., Bangalore Suresh N.T., Garcia X., Llorente O., Estrada L., Acuna V., Birk S., 2025. Guidance Document – Cost-Benefit-Analysis in freshwater ecosystem restoration. EU H2020 research and innovation project MERLIN deliverable D3.4. 114 pp. https://project-merlin.eu/outcomes/deliverables.html Due date of deliverable: 30/09/2025 Actual submission date: 30/09/2025 Revised version, submission date: 22/10/2025 MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 3 MERLIN Key messages 1. Cost-Benefit Analysis (CBA) is essential for evidencebased freshwater restoration planning. CBA can support both public and private investment. 2. Freshwater ecosystem restoration delivers multiple co-benefits (or ecosystem services). 3. Valuing ecosystem services in a freshwater context requires tailored approaches. 4. This guidance complements existing CBA frameworks, focusing on how to assess the costs and benefits of freshwater ecosystem restoration. 5. Seven ecosystem services are included: biomass provision; flood risk mitigation; carbon sequestration; nutrient retention; drought mitigation; recreation and habitat provision. 6. A coupled biophysical – economic modelling tool developed by MERLIN enables users to assess the effects of restoration on flood risk mitigation and nutrient retention using EU-wide datasets 7. Real-world case studies - including the use of CBA on 5 MERLIN Case Studies - are presented to show how the methods work in practice. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 4 Acronyms Acronyms BEAM Basic European Asset Map CBA Cost Benefit analysis CS Case Study CO2e Carbon dioxide equivalent DR Discount rate EA Economic Appraisal EAD Expected Annual Damage EEA European Environment Agency ES Ecosystem Service ETS Emissions Trading System EU European Union FWE Fresh Water Ecosystem GDP Gross Domestic Product GHG Greenhouse Gas GWP Global Warming Potential IRR Internal Rate of Return LAWA Länderarbeitsgemeinschaft Wasser (German Working Group on Water Issues) N Nitrogen NPV Net Present Value P Phosphorus SCC Social Cost of Carbon SP Stated Preference SWAT Soil and Water Assessment Tool VCMS Voluntary Carbon Market System WTA Willingness to accept WTP Willingness to pay MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 5 MERLIN Executive Summary Applying Cost Benefit Analysis to support freshwater ecosystems restoration This guidance document provides targeted support for applying Cost-Benefit Analysis (CBA) in the planning and assessment of freshwater ecosystems restoration. CBA is widely used across the European Union to justify public investments, particularly under legal frameworks such as the Floods and Water Framework Directive and Nature Restoration Regulation. Applying CBA to freshwater restoration is challenging due to data gaps, uncertainty, and the complexity of valuing ecological and social cobenefits. This guidance offers practical methods for integrating ecosystem service (ES) values into CBA. It covers flood and drought risk mitigation, nutrient retention, carbon sequestration, recreation, habitat provision and is supported by five MERLIN case studies and the German AMBERS project. It details how to identify, quantify, and monetise key ES benefits using biophysical modelling (notably SWAT+) and economic valuation (sections 4-12). We provide starting points for the estimation of restoration costs (Section 3), and discuss how CBA can inform the development of private co-funding of restoration (Section 13). The guidance is intended for practitioners conducting CBAs, as well as decision-makers and project developers commissioning or interpreting them. It helps users understand what CBA can offer, how to structure it effectively, and how to use it to support transparent, evidence-based planning and investment decisions. This is not a comprehensive manual for conducting CBAs but a practical resource for incorporating ecosystem services into CBA and should be used alongside generic CBA guidance documents. Estimating restoration costs Costs of restoration generally include investment/construction costs, land acquisition or opportunity costs, and recurring costs. Although literature-based estimates or unit costs can be used in strategic planning phase, these are often scarce or inconsistent for freshwater restoration measures: When available, site-specific estimates are preferable, especially considering the high sensitivity of costs to context-specific factors (such as land price and accessibility). Quantifying and monetising ecosystem services Restoration generates a wide range of ecological, social, and economic effects. In a freshwater restoration context, the ecosystem services (ES) framework is often used to identify and describe benefits from interventions in natural ecosystems. There are three main categories of ecosystem services: provisioning (e.g. water supply, food production), regulating (e.g. water purification, flood control) and cultural (e.g. recreation, aesthetics). Quantifying impacts of restoration on ecosystem services involves linking restoration actions to ecological responses and societal benefits, often using hydrological or biophysical models. Ecosystem service valuation methods are broadly divided into preference-based and market-based approaches. Preference-based methods aim to capture the value individuals place on ecosystem services based on their preferences. They are particularly useful for valuing non-market services such as cultural or recreational benefits. Marketbased methods use existing price data or cost proxies and are best suited to provisioning services. Flood risk mitigation Freshwater ecosystems help reduce flood risks by storing and gradually releasing rainwater, which lowers run-off and flattens peak river flows. Restoration measures that expand or enhance these ecosystems provide significant flood risk mitigation benefits. These benefits can be valued either by estimating the costs of engineered flood protection that would otherwise be needed – known as replacement costs, or by calculating the damages avoided thanks to restoration – known as avoided damage costs. The avoided damage approach is best suited to areas with frequent flooding and little existing infrastructure, while the replacement cost approach is more appropriate in highly regulated systems where restoration can reduce the need for costly flood protection upgrades. Climate change mitigation Freshwater ecosystems are important carbon sinks, storing carbon in lakes, reservoirs, and wetland vegetation. The carbon sequestration potential of restoration efforts can be quantified by comparing carbon sequestration in baseline and restored scenarios, converting these to CO₂ equivalents, and assigning a monetary value per ton of CO2e. Valuation approaches include emission trading prices, the social cost of carbon, or efficient CO₂ prices, with the choice depending on the purpose MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 6 and national guidelines. Market prices should generally be avoided in CBAs, as they do not reflect the full societal cost of emissions. Most EU or member state CBA guidance recommends using social cost of carbon or efficient prices. Nutrient retention Freshwater ecosystems help reduce nutrient pollution by retaining and removing excess nutrients through plant uptake, sediment storage and denitrification. Accurately valuing these benefits requires robust modelling, as nutrient retention depends on complex interactions in both land and water. For economic valuation, the replacement cost method — estimating what it would cost to achieve the same nutrient removal with constructed wetlands — can be used, or environmental damage costs. When using a replacement cost method, one should ensure there is societal demand for the replacement and that replacement is the least-cost alternative. Where possible, it is preferable to use national data on damage costs for more context-specific and reliable results. Drought mitigation Peatland and wetland restoration enhances a landscape’s ability to absorb and slowly release water, improving soil moisture, supporting higher river baseflows, and boosting groundwater and reservoir recharge. Increased water availability during drought benefits hydropower generation (by stabilising low and high flows and reducing reservoir sedimentation), irrigated agriculture, and drinking water supply. The economic value of these benefits can be estimated using production functions, market prices, or cost-based methods, depending on the sector. Supporting river baseflow also helps maintain navigable water levels for shipping and supports water-based recreation, both of which are sensitive to water levels. For accurate valuation, localised assessments are essential, as drought impacts and recreational benefits are highly site-specific. However, impacts on irrigated crops can often be valued using standard crop and revenue data. Recreation Wetlands, floodplains, and rivers provide diverse recreation opportunities, from fishing and wildlife watching to hiking, cycling, and water sports. Restoration can boost recreation by increasing species diversity, improving water quality, and expanding accessible natural areas. The economic value of these benefits is typically estimated using preference-based or benefit transfer methods, based on either increased recreation capacity or the balance of supply and local demand. As more recreation opportunities do not necessarily lead to higher use — benefits are greatest where unmet demand exists – it is preferable to take demand into account. Large-scale restoration can also attract visitors from outside the local area, so preferably both local and regional recreation dynamics are included in CBA. Habitat provision Habitat provision reflects the availability and quality of habitats in rivers and floodplains, underpinning biodiversity and supporting a range of regulating ecosystem services. Restoration measures like rewetting peatlands and reconnecting floodplains can significantly enhance habitat quality and diversity. Quantifying habitat provision relies on indicators such as habitat area, connectivity, and species composition: as there is no internationally agreed best practice, indicators and modelling approaches are often tailored to the local context. Economic valuation typically uses non-market approaches, including preferencebased methods or benefit transfer, to estimate the non-use value people place on biodiversity. Costbased methods can also be used to estimate total restoration value but should not be combined with other ES valuations to avoid double counting. Biomass provision Freshwater ecosystem restoration can affect biomass provision by changing land use and/ or biophysical conditions, influencing the production of food, fibre, fodder, or energy crops. Biomass provision is best quantified as harvested yield (e.g., tons per hectare), using national or EU agricultural statistics or field data. For social CBA, it is recommended to value biomass provision using total yields multiplied by market prices. Methods that separate ecosystem and human contributions — such as production functions, land rental prices, or resource rents – can be used for assessing impacts on the farming sector specifically. Connecting CBA to private benefits The main goal of a CBA is to assess whether restoration benefits society as a whole, not just specific sectors. However, by identifying and quantifying a broad range of societal benefits, a CBA can also inform private co-funding opportunities — such as those linked to markets for biomass, recreation, or carbon credits. To inform private actors’ business cases, costs and benefits have to be attributed to specific stakeholders, and valuation methods that reflect actual revenues or cost savings should be used. Other key differences between CBA and business cases include the timing and time horizon of MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 7 benefits, and the level of detail included in the cost assessment. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 8 Content 1 Introduction ................................................................................................... 10 1.1 Cost benefit analysis: what and why ....................................................... 10 1.2 CBA to support freshwater investment planning in the EU ................ 10 1.3 Key target audience ....................................................................................... 11 1.4 What is in this guidance? ............................................................................. 11 2 Cost-benefit analysis in freshwater restoration ....................................... 13 2.1 The CBA methodology in general .............................................................. 13 2.2 Challenges and solutions for CBA in freshwater ecosystems restoration ............................................................................................................... 13 2.3 Final results – discounting and appraisal horizon ................................ 14 3 Cost assessment ........................................................................................... 16 4 Quantifying and monetising benefits ......................................................... 19 4.1 Identification & quantification ................................................................... 19 4.2 Co-benefits; valuing ecosystem services monetary valuation .......... 20 5 Stated preference methods and value transfer to value ESs bundles .. 23 6 Flood risk mitigation ..................................................................................... 25 6.1 ES logic chain ................................................................................................ 25 6.2 Avoided damage costs approach ............................................................. 25 6.3 Replacement costs approach ................................................................... 26 7 Climate change mitigation ........................................................................... 30 7.1 ES logic chain ................................................................................................ 30 7.2 Ecosystem service quantification ............................................................ 30 7.3 Ecosystem service valuation ...................................................................... 31 7.3.1 Social cost of carbon ....................................................................... 31 7.3.2 Efficient CO2 prices .......................................................................... 32 7.3.3 Compliance and voluntary market prices ................................... 33 8 Nutrient retention ......................................................................................... 35 8.1 ES logic chain ................................................................................................ 35 8.2 ES quantification .......................................................................................... 35 8.3 ES valuation .................................................................................................. 35 8.3.1 MERLIN workflow valuation method ........................................... 35 8.3.2 Other valuation methods ................................................................ 36 9 Drought mitigation ........................................................................................ 39 9.1 ES logic chain ................................................................................................ 39 9.2 ES quantification .......................................................................................... 39 9.3 ES valuation: Water availability for crop production ........................... 40 MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 9 9.4 ES valuation: Hydropower - sediment control and base flow .......... 40 9.5 ES valuation: Inland shipping ..................................................................... 41 9.6 ES valuation: Water-based recreation on lakes and inland waterways ............................................................................................................... 42 10 Recreation ...................................................................................................... 44 10.1 ES logic chain ................................................................................................ 44 10.2 Ecosystem service quantification & valuation ....................................... 44 10.2.1 Hiking and cycling in reconnected floodplains – confronting supply and demand ..................................................................................... 44 10.2.2 Angling in reconnected floodplains .............................................. 45 10.2.3 Water-based recreation on re-connected river stretch .......... 45 11 Habitat provision ........................................................................................... 47 11.1 ES logic chain ................................................................................................ 47 11.2 ES quantification .......................................................................................... 47 11.3 ES valuation .................................................................................................. 49 11.3.1 Stated preferences .......................................................................... 49 11.3.2 Restoration and Compensation Cost Approaches .................... 50 12 Biomass provision ......................................................................................... 53 12.1 ES logic chain ................................................................................................ 53 12.2 ES quantification .......................................................................................... 53 12.3 ES valuation .................................................................................................. 53 13 Connecting a social CBA to private financing of freshwater ecosystems restoration 55 References .............................................................................................................. 57 Cost assessment MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 16 3 Cost assessment Costs of restoration generally include investment/construction costs and (land-) acquisition (one-off) costs, and (recurring) management costs. The costs of restoration depend on many variables, which may be very project or context-specific, e.g. relating to the design of restoration, regional variations in land prices, labour or energy costs. The REFORM project (Ayres et al., 2014) recommends a cost typology for river restoration (Table 1). Table 1 – Restoration costs typology; adapted from (Ayres et al., 2014). Type Description Planning and design costs Costs incurred during project preparation, including project team set-up, data collection, set project objectives, planning and detailed design and implementation strategy. Transaction costs Costs other than investment or operation and maintenance costs incurred during planning and implementation phase, including for communication, legal fees, decision making processes, quality control. Investment/Construction costs Costs needed to physically implement the project, including e.g. labour, material and equipment, soil sanitation. Land acquisition costs Costs of land or property that needs to be acquired to implement the project – this may also take the form of compensation for current tenants. Maintenance & monitoring costs Annual costs needed for upkeep and repair of the restored area. This may also include monitoring costs to analyse changes in ecological and hydromorphic conditions. The cost categories in bold are the most used in cost-benefit analysis. Instead of land acquisition costs, opportunity costs are often used, i.e. the foregone benefits that would have been received if land use was not changed (e.g. agricultural revenues). Table 2 presents rough cost estimates from MERLIN Case Studies (CS) for some freshwater ecosystems (FWE) restoration measures. Box 3.1 and Box 3.2 provide examples of detailed analysis of restoration costs in two MERLIN CS. A detailed analysis of restoration costs incurred by all MERLIN CSs will be made available in Deliverable 2.5. Table 2 - Implementation costs of restoration in a selection of MERLIN case studies. Type of measures Type of cost €/stretch (km) €/ha Case Studies Riparian buffer Opportunity cost 2,526 MERLIN CS13. Sorraia floodplain Peatland rewetting Land works, ash fertilization, trees sowing, bog vegetation restoration, duckboards building 3,100 MERLIN CS14. Komppasuo peat extraction area Peatland rewetting Blocking of ditches, reprofiling of exposed peat 2,100 MERLIN CS17. Forth catchment River channel restoration and floodplain reconnection Ditches locking, scrapes creation, channels diversion, leaky dams, tree planting. 2,540 MERLIN CS17. Forth catchment Floodplain reconnection Dike setback (large river) 29 M (km) MERLIN CS04. Room for the Rhine branches Floodplain reconnection Side channel (large river) 27 M (km) MERLIN CS04. Room for the Rhine branches Cost assessment MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 17 Lessons learned & recommendations Ø Cost data availability and granularity: Cost data for nature-based solutions and restoration measures is often scarce or inconsistent. At the catchment scale, literature-based unit-cost estimates can be used for strategic planning CBAs. However, at a more local planning and design scale it is advisable to obtain sitespecific data: as illustrated by the case in the Netherlands (Box 3.2), net costs for floodplain reconnection varied significantly across river branches, influenced by factors such as urbanization, land prices, accessibility, and material availability. Site specific data can be obtained by commissioning cost assessments by engineering firms or consulting local contractors. Ø Maintenance and monitoring are long-term commitments: Restoration is not a one-off investment. Longterm costs for upkeep and ecological monitoring should be included in cost estimates. Ø Use of standardized cost typologies improves comparability: Applying frameworks like the REFORM cost typology helps structure cost data and facilitates comparison across projects and regions. Further reading: restoration costs assessment Ø REFORM D1.4: Inventory of restoration costs and benefits D5.2: Cost effectiveness of river restoration Ø Glenk et al., 2022: the cost of peatland restoration, data and analysis. Ø Moxey and Moran (2014): UK peatland restoration: Some economic arithmetic Ø Aerts (2018): a review of cost estimates for flood adaptation. Box 3.1: Estimating opportunity costs of riparian buffer zones in the Sorraia floodplain, Portugal (MERLIN CS 13) Based on spatial analysis using buffer zones around streams (ARCGIS), the surface area of cropland to be converted to buffer zones was determined (860 ha). To break this down to specific crop types, two sources were used - generic crop data from the Portuguese COS2018 land use database (DG Territorio, 2018), and, where available, more detailed crop data provided by the local water authority (In absence of national or local data, the widely available CORINE database can be used). The resulting summarized areas per crop type were then connected to Standard Output Coefficients, the average monetary value of agricultural output at farm-gate price, in €/ha for specific crops (2023 price level; averaged for the Alentejo region in Portugal; derived from EU's Integrated Farm Statistics database https://circabc.europa.eu/ui/group/a9c5638c-8940-4e25-b6de-02ace3e161e7/library/23467ccb-9d33-4376a877-95925673044b). Annual opportunity costs were then discounted over a horizon of 100 years, at discount rate 3%. As a result, the annual production loss related to installing riparian buffer strips in the Sorraia watershed is estimated at € 2.17 million (average €2526/ha), which accumulates to a present value of €70.8 million. Cost assessment MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 18 Box 3.2: Floodplain reconnection in the Netherlands: cost breakdown (MERLIN CS 04) In the preparation phase of the Dutch 'Room for the River' Programme (2010-2020), a database of cost estimates for 1166 river widening measures was developed to support strategy design (Prins & Levelt 2015). This database can serve as input to derive average unit implementation costs (Table 3) and provide information on main cost elements (Figure 3). Table 3 - Unit costs per measure type along the Rhine Branches in the Netherlands per river kilometre (stretch) in MEUR at price level 2023, including VAT (21%). Estimates include CAPEX and OPEX but exclude land acquisition costs. From Kok et al (submitted). Mesure/river stretch Upper Rhine, Waal and Merwede Pannerdensch Kanaal and IJssel Nederrijn/ Lek Floodplain reconnection project (e.g. lowering) 30 12 12 Dike relocation 39 18 32 Construction of side channel or flood channel 22 38 21 Differences in unit costs per river branch can be due to factors such as the degree of urbanisation in the area. Overall, land acquisition (including right of superficies – right to own constructions on land owned by another person or entity) and compensation and physical works make up a large part of the costs. Costs for both land acquisition/compensation and physical works increase with urbanisation. Expected operation and maintenance costs were also derived from the cost database: 0.5% for floodplain reconnection projects and side channels, and 0.4% for dike relocation projects. Figure 3 - Cost breakdown based on cost estimates of Room for the River database. From Kok et al (submitted) Quantifying and monetising benefits MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 19 4 Quantifying and monetising benefits Restoration generates a wide range of ecological, social, and economic effects. Quantifying and monetising these effects or benefits, is the basis of a cost-benefit analyses. In a freshwater restoration context, the ecosystem services (ES) framework is often used to identify and describe benefits from interventions in natural ecosystems. This chapter outlines how freshwater restoration measures affect ecosystem services, and describes the process of identifying, quantifying, and valuing these benefits, referring to tools, existing databases, and relevant projects across Europe. In most cases, the valuation of ecosystems restoration builds on the concept of ecosystem services (HainesYoung & Potschin, 2013), which link ecosystems to human well-being and the economy. Ecosystems provide a large range of services to people, categorised as provisioning services (e.g. water supply, biomass production), regulating services (e.g. climate regulation, flood risk mitigation) and cultural services (e.g. recreation, aesthetic value). 4.1 Identification & quantification Freshwater ecosystem restoration measures can influence a wide array of ecosystem services, both positively and negatively. Identifying and selecting the most relevant benefits for evaluation is the first step. This requires understanding of how proposed interventions affect ecosystem services and which impacts are most critical for local stakeholders and policy objectives. Different types of restoration measures influence ecosystem services in various ways. For instance, rewetting a floodplain may reduce provisioning services like crop production, while enhancing regulating services such as CO₂ sequestration, nutrient retention and flood mitigation. A good starting point to assess which ES are affected by the proposed interventions is to use matrices such as developed by Natural Water Retention Measures (NWRM) (https://www.nwrm.eu/catalogue-nwrm/benefit-tables) or the RESI project (Hornung et al., 2019) which visualise how interventions may positively or negatively impact the delivery of ES (Figure 4). Impacts in these matrices are based on existing experiences and literature (see e.g. Birk and al. 2025); of course, the relationship between restoration and a specific ES may vary from site to site, and such generic matrixes should be used with caution. Figure 4 - Benefit table of freshwater ecosystems restoration measures and their likely effects on ES – adapted from nwrm.eu It is important to consider not only the benefits of restoration, but also potential ecosystem disservices (EDS)— unintended negative impacts on human well-being, such as increased mosquitoes, ticks, or allergies from restored wetlands (Guo et al., 2022). Including these in assessments helps identify locally adapted, widely supported solutions. Expanding current evaluation methods and cost-benefit analyses to include both positive Quantifying and monetising benefits MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 20 and negative effects can ensure comprehensive, integrated planning processes and increase public support (Guo et al., 2022; Potgieter et al., 2019; Roy et al., 2012). Before the impacts of proposed interventions can be monetised for the cost-benefit analysis, these impacts must be quantified. This involves tracing the links between restoration actions, ecological responses, and the benefits to society and nature. Quantification of the impact of freshwater ecosystems restoration interventions often requires hydrological and/ or biophysical modelling, using models and tools such as the hydrological SWAT+ and/or ES modelling tools such as InVEST (Natural Capital Project Data Hub | Natural Capital Project). In chapters 5-12 we discuss quantification approaches for specific ecosystem services in more detail. 4.2 Co-benefits; valuing ecosystem services monetary valuation There are several types of methods to value ecosystem services in monetary terms, each with different levels of precision, data needs, and applicability. Broadly, valuation methods fall into two categories: preferencebased methods and market-based methods. Preference-based methods aim to capture the value individuals place on ecosystem services based on their preferences. They are particularly useful for valuing non-market services such as cultural or recreational benefits. Market-based methods use existing market data or cost proxies to estimate the ecosystem service values – this approach is generally more straightforward but may not capture full welfare impacts. They are best suited to provisioning services. We shortly introduce main valuation methods for ecosystem services and main underlying assumptions. Revealed Preference Methods estimate the value of an ecosystem service based on observed behaviour in actual markets or real-world decisions. Often used approaches include: Ø The Travel Cost Method (TCM), in which the value of recreational sites is determined by analyzing how much people spend to visit them (both in time and money). This captures the consumer surplus from recreational activity, which reflects individual welfare gains. The underlying assumption is that the travel costs are an accurate reflection of the willingness to pay for the activity – this may be an over or underestimation (e.g. if people are also visiting other things; or would be willing to pay more). Ø Hedonic Pricing, which involves evaluation of how environmental attributes (e.g. proximity to water bodies, good or bad water quality) influence property prices. This measures the marginal willingness to pay for environmental attributes. The underlying assumption is that markets are competitive, and buyers are fully informed – if this is not the case, the prices may be underestimated. Stated Preference Methods estimate the value of an ecosystem service by directly asking individuals about their preferences in hypothetical scenarios. An advantage of this approach is that it can be used to capture non-use values such as the bequest/existence value of biodiversity. Often used approaches include: Ø Contingent Valuation (CV), in which individuals are asked directly about their willingness to pay (WTP) for specific ecosystem services or restoration outcomes. Respondents are assumed to understand and truthfully respond to questions – as such, the approach is quite sensitive to survey design and biases, which may lead to under or overestimation of the WTP. Ø Choice Experiments (CE) present respondents with different scenarios involving trade-offs, allowing estimation of marginal WTP for individual service attributes (e.g. 10% increase in water quality, biodiversity, recreational opportunities). The underlying assumption is that respondents understand and evaluate tradeoffs well – this requires careful experimental design. Similar to contingent valuation, biases may still apply though the risk is lower with choice experiments. Market-Based Methods use existing market data or cost proxies to estimate ES value. Often used approaches include: Ø The Market Price approach uses actual prices of goods and services (e.g. fish, timber) to value provisioning services. The underlying assumption is that markets are efficient and stable, and that prices reflect true value – meaning that the price indeed reflects the willingness to pay for a service, and there are no significant externalities. Ø Production Function approaches estimate how ES contributes to economic outputs (e.g. how pollinators affect crop yields). This requires a clear, well-understood causal relationship between the ES and production of the marketed good: a detailed production function as well as detailed models and data to estimate the impact are required and, in practice, are often unavailable. Cost-Based Methods estimate the value of services based on the costs that would be incurred if those services were lost or had to be replaced by human-made alternatives. These methods do not measure people’s preferences directly and but can be useful if preference-based data is unavailable or too complex to derive. Often used approaches include: Quantifying and monetising benefits MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 21 Ø The Replacement Cost approach, which estimates the cost of replacing an ES with man-made alternatives (e.g. a wetland can be replaced with a water treatment plant). The underlying assumption is that artificial substitutes can fully and efficiently replace ecosystem services, and that this would indeed be done if services were lost. Values may be underestimated (e.g. if a service is only partially substitutable) or overestimated (if replacement costs are inflated or over dimensioned). Ø The Avoided Damage cost approach, which values services based on the costs they help prevent (e.g. flood regulation reducing property damage). The underlying assumption is that it is possible to accurately estimate avoided damage – and that this is a good reflection of individual WTP (disregarding the risk appetite of individuals). If primary valuation studies are not feasible due to time, budget, or data constraints, benefit transfer methods offer a practical alternative, which requires limited resources. These methods estimate the value of ecosystem services by transferring results from existing studies conducted in similar contexts. There are two main types of benefit transfer: Ø Unit Value Transfer, in which a single value (e.g., € per hectare) from a different study site is used in the evaluation. Ø Benefit transfer functions, often based on a meta-regression model analyzing a wide range of values from different study sites, which allows for accounting for site-specific variables. Both these approaches are sensitive to transfer errors – if sites, interventions or population differ, values can be over – or underestimated. States preference and benefit transfer methods may be applied to value a bundle of ESs provided by freshwater ecosystem restoration, but this was not tested in MERLIN. Nevertheless, this approach may be relevant in given contexts and is therefore briefly discussed in Section 5. In Chapters 6-12 we present and discuss monetary valuation approaches for specific ecosystem services in more detail. Note that there is an important difference in valuing ecosystem services in the context of CBA and in the context of ecosystem accounting (e.g., following the System of Environmental Economic Accounting or SEEA). In the first case, welfare values are used, that include changes in the value of both consumer and producer surpluses resulting from a project or policy. In the latter case, exchange values are used, that do not include consumer surplus and where values are consistent with the values found in national accounts. This report focusses on welfare based valuation, whereas the SEEA Ecosystem Accounting (United Nations, 2024) provides more insights in the valuation methods relevant for ecosystem accounting. Quantifying and monetising benefits MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 22 Further reading: Ø Identification and quantification of restoration benefits Ø Benefit tables – linking ecosystem services and measures: o NWRM benefit tables o Hornung et al. (2019) Linking ecosystem services and measures in river and floodplain management Ø Vermaat et al. (2013) REFORM D2.3 Valuing the ecosystem services provided by European river corridors – an analytical framework; includes a list of ecosystem services potentially provided by river corridors, alongside key abiotic and biotic conditions, relevant scale and societal beneficiaries. Ø Grizetti et al. (2015) EU cookbook for watershed-scale ecosystem service indicators for rivers Ø RESI River Project (Germany): Various studies on spatial tools and matrices for ES trade-offs ((Hornung et al., 2019; Podschun et al., 2018; Stammel et al., 2021). Ø United Nations (2022). Guidelines on Biophysical Modelling for Ecosystem Accounting. United Nations Department of Economic and Social Affairs, Statistics Division, New York. Ø MERLIN Deliverable 3.3: The MERLIN modelling workflow to assess the biophysical and economic impact of freshwater ecosystem restoration at catchment scale Ø Monetary valuation of restoration benefits Ø UNSystem of Environmental Economic Accounting (UN et al., 2024) Ø NCAVES and MAIA (2022). Monetary valuation of ecosystem services and ecosystem assets for ecosystem accounting: Interim Version 1st edition. United Nations Department of Economic and Social Affairs, Statistics Division, New York. Ø Vermaat et al. (2016) Assessing the societal benefits of river restoration using the ecosystem services approach Ø DEFRA (2020) Enabling a Natural Capital Approach. Data sources, tools and studies for natural habitats and environmental impacts including economic valuation evidence for the UK, intended to facilitate natural capital assessment following the HM Treasury Green Book guidance. Stated preference methods and value transfer to value ESs bundles MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 23 5 Stated preference methods and value transfer to value ESs bundles This section addresses stated preference and value transfer methods to estimate the value of a bundle of ESs provided by freshwater ecosystems restoration. In evaluating the societal benefits of freshwater ecosystem restoration, stated preference methods and value transfer approaches offer distinct advantages and limitations, especially when applied within spatially explicit cost-benefit analyses. Stated preference methods, such as contingent valuation and choice experiments, are particularly valuable because they can capture both use and non-use values of ecosystem services—ranging from recreational benefits to cultural and existence values. These methods allow researchers to directly estimate people’s willingness to pay (WTP) for specific restoration outcomes, which is essential for assessing the full societal value of non-market services. Examples of studies that integrate all ESs in one SP survey in the context of freshwater ecosystems can be found in Logar et al., (2019), De Blaeij et al., (2009), Kourtis & Tsihrintzis (2017) and Saarikoski et al., (2022). Moreover, when carefully designed, SP studies can be tailored to reflect stakeholder preferences and policy-relevant trade-offs.Click or tap here to enter text. However, stated preference studies are often resource-intensive and complex to implement. They typically require extensive survey design, piloting, and statistical analysis, which can be a barrier in large-scale or multisite assessments. Another challenge lies in the interpretation of results: respondents often value bundles of ecosystem services rather than individual ones, making it difficult to isolate the value of specific services. Furthermore, these studies are rarely spatially explicit, which limits their usefulness in modelling geographically differentiated restoration scenarios. Unless the study is designed to link WTP to incremental biophysical changes, integrating results with ecological models can be problematic. Value transfer, on the other hand, offers a more practical and cost-effective alternative, especially when primary data collection is not feasible. By using results from existing valuation studies, analysts can estimate benefits across multiple sites or scenarios with relatively low effort. This approach is particularly useful for initial screening-level CBAs or when broad coverage is needed. Recent advances in meta-analytic value transfer methods have improved the ability to adjust for contextual differences, making the approach more robust. Several meta-analytic value transfer functions exist for ESs provided by floodplains (Perosa et al., 2021), wetlands (Brander et al., 2006; Ghermandi et al., 2010; Eric et al., 2022), lakes (Reynaud & Lanzanova, 2017) and rivers (Brouwer & Sheremet, 2017). However, to our knowledge, only one of these value transfer functions directly model the economic value of ecosystems restoration (Brouwer & Sheremet, 2017). Other functions model the ESs values provided by freshwater ecosystems. Nevertheless, these functions may be used to estimate restoration benefits in cases when they show a relationship between ESs values and site characteristics on which restoration has an impact, such as the extent of freshwater ecosystems (e.g. wetland area) and/or freshwater ecosystems condition (e.g. chemical and ecological status). Yet, value transfer also comes with limitations. Transferred values may not accurately reflect local ecological, cultural, or economic conditions, especially in novel or highly site-specific restoration contexts. The lack of precision can be problematic when detailed, spatially explicit valuation is required. Moreover, without careful calibration, there is a risk of misrepresenting actual benefits, which can undermine the credibility of the analysis. For instance, when value transfer functions are developed from global meta-analysis, they may not provide accurate estimates at national or local scale (Natho & Hudson, 2024). To address these challenges, it is increasingly recommended to integrate economic valuation approaches with biophysical models. Spatially explicit modelling enhances the robustness of ecological predictions, captures geographic and temporal variation in ecosystem service flows, and supports more nuanced scenario analysis. This integration is particularly important for freshwater restoration, where ecosystem service provision is closely tied to hydrological and land-use dynamics. In summary, while stated preference methods offer depth and richness in capturing societal values, they require careful design and are less suited to spatially explicit modelling unless integrated with ecological data. Value transfer is more scalable and efficient but must be applied with caution to ensure contextual relevance. A hybrid approach—combining targeted primary valuation with broader value transfer—may offer the best balance between accuracy and practicality in freshwater restoration CBAs. Stated preference methods and value transfer to value ESs bundles MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 24 Further reading: Ø Johnston et al. (2017) - Contemporary Guidance for Stated Preference Studies Ø Johnston et al. (2021) – Guidance to enhance the validity and credibility of environmental benefit transfers. Ø Johnston et al. (2015) – Benefit transfer of environmental and resource values: a guide for researchers and practitioners Flood risk mitigation MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 25 6 Flood risk mitigation 6.1 ES logic chain Ecosystems in freshwater catchments play a crucial role in reducing flood risks through their capacity to store and gradually release rainwater, reducing run-off and flattening peaks of high river flows. FWE restoration measures that increase the extent and/or capacity of ecosystems regulating water flows deliver flood risk mitigation benefits to property owners and residents living in flood prone areas of river catchments (Table 4). Table 4 – Flood risk mitigation ES logic chain. Ecosystem types Factors determining supply Factors determining use Potential physical metric(s) Benefits Main users and beneficiaries Ecological Societal Terrestrial and freshwater ecosystems (especially floodplains) within river catchments Extent and condition of vegetation, ambient climate factors Land use, ecosystem management Existing flood protection infrastructures (e.g. dikes, levees) Change in flood retention volume (m3)/dike height (m) required for flood protection Avoided flood protection costs Property owners and residents – households, businesses, insurance sector, government Land use in flood-prone areas Number/area of residential/busin ess/farm properties at flood risk in case of a 100year flood event Avoided damages of flood events Flood risk mitigation benefits can be monetised through two valuation approaches: Ø Replacement costs: the costs that would have been spent on flood protection infrastructure to reach the same level of flood mitigation had the restoration not taken place, or Ø Avoided damage costs: or prevented (expected) damage costs caused by floods that would have occurred had the restoration not taken place. Whether to use a replacement cost or an avoided damage costs approach depends on the biophysical and socio-economic context of the valuation. For instance, in the Dutch Rhine River basin (MERLIN CS 04), water flows are highly regulated, and flood barriers are regularly upgraded to prevent river flooding damages, at a safety standard (e.g. 1:10.000 year) set in law. Therefore, benefits of FWE restoration will materialise as a reduction of infrastructure works required to maintain flood protection standards: in this case, the replacement cost approach should be used to monetise benefits. In the case of many large rivers with an (originally) flood-prone (e.g. low-lying and flat) hinterland, such protection infrastructure is present. Conversely, many small river catchments do not have an extensive flood protection system and are subject to regular flood events. In these catchments, the avoided damage cost approach is best suited to monetise benefits. The rest of this section discusses possible quantification and valuation methods, with existing models and datasets, for the avoided damage costs approach (Section 6.2) and the replacement costs approach (Section 6.3), embedding lessons learnt from MERLIN CS CBAs, and end with recommendations. 6.2 Avoided damage costs approach Estimating flood damage costs implies to quantify (1) flood hazard, i.e. the probability of flood occurrence, (2) flood exposure, i.e. the assets value exposed to flood hazard and (3) flood vulnerability, i.e. the proportion of assets value at loss in case of flooding. Flood hazard Flood hazard can be quantified with return period-based flood extent and depth maps (probability). For instance, a map with a return period of 1:100-year event shows the extent and depth of floods that occur every 100 years, corresponding to a probability of occurrence of 1%. A range of return periods are covered to quantify the effects of flood events with different magnitudes and frequencies. For instance, in Poland, the State Water Management Company published flood extent and depth maps with return periods of 10, 100 and 500 years. Climate change mitigation MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 32 7.3.2 Efficient CO2 prices Efficient prices reflect the costs needed to achieve emission reductions at the lowest possible expense. As the deadline of emission reduction targets approaches, marginal costs increase due to the exhaustion of low-cost options. Since the European Emission Trading System (ETS) does not cover all economic actors (yet) or the full volume of GHG emissions, additional policy measures are implemented outside the ETS in order to reach reduction targets. Efficient price paths embed both the ETS and additional policies. Usually, prices are calculated for various scenarios, including e.g. economic scenarios and scenarios for the ambition level of emission reduction targets (see Box 7.4). Box 7.3: Use of Social cost of Carbon in German policy evaluation The German Federal Environment Agency (UBA) applies the social cost of carbon (SCC) as a central metric in environmental policy evaluation (Matthey et al., 2024). For the reference year 2021, the UBA set the SCC at €282.50 per tonne of CO₂, based on a 1% pure time preference rate. This value was derived from interpolated data between 2020 and 2023 (Matthey et al., 2024). The UBA also applies equity weights, to account for global income disparities and intergenerational equity. Thus, when a 0%-time preference rate is applied, treating future generations equally, the SCC rises significantly to €880 per tonne of CO₂. Table 8 Social cost of carbon. From Gesellschaflichte Kosten von Umweltbelastungen (Matthey et al., 2024) Costs in Euro (2024) per ton CO2 2024 2030 2050 1% time preference rate 300 335 435 0% time preference rate 880 940 1080 Climate change mitigation MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 33 7.3.3 Compliance and voluntary market prices Driven by targets set on emissions of greenhouse gases, the EU set up a cap-and-trade system, the Emissions Trading System (EU ETS). The ETS covers emissions from energy-intensive industries, power and heat generation, commercial aviation, and maritime transport. Starting in 2027, it will expand to include smaller industrial sectors, buildings, and road transportation. Carbon pricing is not unique to the EU: similar systems exist globally. Within the EU, prices are projected to rise from €80 per ton/CO2 equivalent today to € 145 by 2030 and €177 by 2035 (Tiseo, 2025). In parallel to compliance markets, voluntary carbon markets (VCM) have emerged. These markets allow individuals, companies and organisations to buy carbon credits to offset their emissions, often motivated by corporate sustainability goals or to compensate for specific activities such air travel. Unlike the compliance markets, VCMs are not governed by legally binding emissions caps. In the voluntary markets, projects which capture and/ or sequester CO2 - such as reforestation, renewable energy - can obtain credits verified by independent third parties, which can then be sold to buyers. Despite rapid growth of the market in the past decade (e.g. supported by the Carbon Removals and Carbon Farming Regulation, EU 2024), prices in the voluntary carbon market are quite volatile. Between 2017 and 2021, prices ranged from (dollar) 2 to (dollar) 4 per ton CO2. Key factors in this price are the type of carbon mitigation projects (forestry and land use have higher prices due to perceived high quality resulting in a nonhomogenous market), the age of the credits (newer credits higher values), investor demand and geopolitical tensions (Liu, 2024). Lessons learned & recommendations Ø For rapid cost-benefit analyses (CBAs), lookup tables can be used to estimate carbon sequestration potential. However, for more accurate assessments — especially at the local scale — it is preferable to use national or regional data on greenhouse gas emissions by ecosystem type. Ø For extended CBA, more detailed modelling of dynamic carbon cycling processes is recommended. Recent literature highlights that river-floodplain systems can act either as carbon sources or sinks, depending on the balance of various carbon cycling processes (Petch et al., 2023). The effects of restoration interventions on greenhouse gas emissions—particularly methane—remain uncertain. For instance, rewetting previously drained agricultural land may lead to a temporary increase in emissions. Box 7.4: Carbon policy evaluation in the Netherlands: efficient CO2 price paths In the Netherlands, efficient prices are the official approach to price CO2 in economic policy appraisals. These efficient prices are derived by Aalbers et al., (2017), using a discount rate of 3.5% (Table 9). In the 'high scenario', there is both high economic growth and a high emission reduction target. In the 'low' scenario there is low economic growth, and a low emission reduction target. The large bandwidths for the reported prices under the 2°C target stem from 1) the definition of the 2-degree goal, i.e. is temporary overshooting of the reduction target allowed or not, and 2) assumptions regarding large-scale applicability, capacity and costs of technologies such as carbon capture and storage and biogas production. Table 9 Efficient prices and ETS prices of 1 ton CO2 (in euros) used in the High and Low scenarios and in in the two degree scenario. Scenario Price type 2015 2030 2050 High Efficient price 48 80 160 ETS price 5 40 160 Low Efficient price 12 20 40 ETS price 5 15 40 2 degree Efficient price 60-300 100-500 200-1000 ETS price 5 100-500 200-1000 Climate change mitigation MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 34 Ø When evaluating the carbon sequestration potential of restoration efforts, vegetation management should be taken into consideration. If biomass is removed from the restored area, the assessment should account for whether the carbon is likely to be emitted or sequestered, depending on its end use (see Box 7.2; Pfau et al. 2019). Ø When monetizing the impacts of carbon sequestration, market prices should be avoided as they do not reflect the full economic cost of emissions. Instead, the social cost of carbon or efficient pricing mechanisms should be used. Many EU countries have established official guidelines for carbon pricing in CBAs (see Box 7.3 & 7.4). Ø An additional assessment using prices from the EU Emissions Trading System (ETS) or voluntary carbon markets can offer valuable insights into the potential revenue streams of a project, particularly if it becomes eligible for carbon credits. Further reading: Ø Penman, J., Gytarski, M., Hiraishi, T., Krug, T., Kruger, D., Pipatti, R., Buendia, L., Miwa, K., et al. (2003). Good Practice Guidance for Land Use, Land-Use Change and Forestry. Intergovernmental Panel on Climate Change, National Greenhouse Gas Inventories Programme (IPCC-NGGIP) Ø IPCC Task Force on National Greenhouse Gas Inventories. 2013 Supplement to the 2006 IPCC Guidelines for National Greenhouse Gas Inventories: Wetlands. (IPCC, 2014). Ø Nordhaus, W.D. (2017). Revising the social cost of carbon. Proceedings of the National Academy of Sciences, 114(7), 1518-1523. DOI: 10.1073/pnas.1609244114. Ø World Bank (2025). State and Trends of Carbon Pricing 2025. https://www.worldbank.org/en/publication/state-and-trends-of-carbon-pricing Nutrient retention MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 35 8 Nutrient retention 8.1 ES logic chain Water purification services are the ecosystem contributions to the restoration and maintenance of the chemical condition of surface water and groundwater bodies through the breakdown or removal of nutrients and other pollutants (United Nations, 2024). This guidance is focused on nutrient retention. This choice was driven by the importance of nutrient pollution for humans and ecosystems, as well as the large body of existing works on nutrient retention modelling and monetary valuation, upon which this guideline could build, while recognising that other pollutants, for instance pesticides, also have large effects on humans and ecosystems. Pollution by nutrients has harmful effects on human use of water, for instance by restricting access to water for recreation activities (Pretty et al., 2003) and human health, for instance in case of nitrates excess in drinking water (Van Grinsven et al., 2010). High concentrations of nutrients in water bodies also lead to eutrophication, with negative consequences on ecosystem health and biodiversity (Grizzetti et al., 2011). By retaining nutrients, ecosystems mitigate the harmful effects of nutrients on humans and ecosystems, thereby providing a broad range of benefits, such as reduced water treatment costs, improved health and reduced eutrophication. Potential beneficiaries of this ES are water supply companies, water-based recreation businesses, water-front property owners and households. Freshwater ecosystems have a high capacity to supply nutrient retention ES. N and P concentration can be reduced by algae and plant uptake, as well as storage in sediments and soils in water bodies and wetlands (Reddy et al., 1999). Moreover, anoxic conditions in wetlands, riparian areas and water bodies stimulate the release of N to the atmosphere through denitrification by bacteria (Seitzinger et al., 2006). Table 10 - Nutrient retention logic chain. Ecosystem types Factors determining supply Factors determining use Potential physical metric(s) Benefits Main users and beneficiaries Ecological Societal Peatlands, wetlands, rivers and lakes Ecosystem condition: chemical state, biological composition Location, type and volume of released water pollutants Demand for clean water for human consumption, water-based recreation, or other uses Tons of pollutants removed by type of pollutant Reduced water treatment costs, improved health outcomes, reduced eutrophication Water supply companies, water-based recreation businesses, water-front property owners, households 8.2 ES quantification Nutrient retention depends on complex climatic, physical, chemical and biological processes which take place in both terrestrial and aquatic ecosystems. Different modelling tools with various levels of complexity exist. InVEST and ESTIMAP, two multi–ES modelling platforms, are relatively easy to use but have important limitations with regards to modelling the effects of FWE restoration. For instance, InVEST does not capture instream retention processes, while ESTIMAP runs at rough scales, providing yearly averages of, on average, 180 km² sub-basins. The SWAT+ model captures both land and in-stream retention and has been widely used to model N retention. It is however requiring specific expertise and is data intensive. To facilitate the use of SWAT+ to model the effects of FWE on N retention, the MERLIN project has developed a user-friendly modelling workflow based on EU-wide publicly available datasets (Garcia et al., 2025). The MERLIN workflow can be used to model the effects on N retention of 5 different restoration measures (peatland rewetting, wetland rewetting, riparian buffers, channel restoration and floodplain reconnection). 8.3 ES valuation 8.3.1 MERLIN workflow valuation method In the modelling workflow developed by MERLIN (see Deliverable report D3.3 and annex 2), a replacement cost approach has been implemented to allocate a monetary value to nutrient retention benefits of ecosystem restoration. This approach values the nutrient retention service provided by the restoration of ecosystems in a water catchment on the assumption that the service would be replaced by constructed wetlands if it did not exist. This approach was adapted from La Notte et al. 2017. Replacement constructed wetlands are Nutrient retention MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 36 dimensioned to provide a nutrient retention service equivalent to the service provided by the restoration of ecosystems. Depending on the context, i.e. the relative importance of N and P retention for enhancing water quality, users may decide to dimension the constructed wetlands based on the N, P or both N and P retention service. The monetary value of the nutrient retention service equals the sum of annualised net present value of the capital investment costs and yearly operational costs of the constructed wetlands. Annex 2 provides a detailed methodological description of the valuation method implemented in the MERLIN tool. The validity of this replacement cost approach hinges on three core assumptions. First, it is assumed that the constructed wetland can provide exactly the same ES as the ecosystems. Constructed wetlands have similar functions than freshwater ecosystems and were dimensioned to provide the same level of nutrient retention. Second, it is assumed that there is an actual societal demand for the building of constructed wetlands as an alternative to the nutrient retention service. This can be checked using the water bodies quality status of the Water Framework Directive. If the WFD water quality targets reflect the societal demand for water quality, we can hypothesise that society would be willing to pay for constructed wetlands only where WFD targets are not met in the water basin of interest (see Box 8.1). Third, it is assumed that constructed wetlands are the least-cost alternative to the nutrient retention service. La Notte et al. (2017) state that constructed wetlands are much less expensive than wastewater treatment plants and that the chosen constructed wetland type, i.e. free water surface constructed wetlands, is the less expensive one. However, other (possibly less costly) alternatives to constructed wetlands could be considered, for instance the reduction of nutrients losses from agricultural land management. Ideally, a replacement cost approach should consider all possible alternatives to the ES and be based on the combination of alternatives that minimise replacement costs. Grossman et al. (2012) provide an example of such a cost-minimisation approach. This kind of approach requires context specific data and models, thus was not feasible to implement in the MERLIN modelling workflow. 8.3.2 Other valuation methods Nutrient retention may also be valued through the environmental damage costs caused by excess of nutrients in water bodies. Nutrients and eutrophication damage costs include drinking water treatment costs to remove nutrients, health costs due to the presence of nitrates in drinking water, reduced value of waterside properties, reduced recreational and amenity value, and loss of biodiversity. Price & Heberling (2018) reviewed empirical literature on the effects of source water quality on drinking water treatment costs. They found that nutrient loads in source water had significant but moderate effects on treatment costs: a 1% increase in nutrient concentration led to 0.06% and 0.02% increase of treatment costs for N and P respectively. The effect of sediment load was higher (0.23%). Note however that those results were based on a handful of studies, mainly from the US. Van Grinsven et al. (2010) estimated that the incidence of colon cancer increased by 3% due to nitrate contamination of groundwater in the EU, corresponding to a cost of 0.7 euro per kg of nitrate N leaching, but their estimate was inferred from one epidemiological study in the US. Social costs of Nitrogen have been estimated in Europe (Brink et al., 2011; Van Grinsven et al., 2013), but these estimates are based on a few studies and have a large uncertainty. In some European countries, national estimates may be available. For example, the German Federal Environment Agency (Matthey et al., 2024) determined the social costs of Nitrogen at €21.44/kg N and €158.22/kg P (price level 2021, per year). Nutrient retention MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 37 Box 8.1: Nutrient retention benefits estimation with the MERLIN workflow - MERLIN CS17 Forth Catchment In the Forth Catchment, the SWAT+ model estimates that peatland rewetting enhances nutrients retention and therefore decreases quantities of nutrients reaching lakes and rivers. The effects were quantified as 88 tons of Nitrogen and 11 tons of Phosphorus removed per year. Applying the MERLIN plugin replacement cost approach the value of the nutrients retention benefit was estimated at 640 K€/year for N retention and 595 K€/year for P retention, i.e. 7.4 €/kg Nitrogen and 55.7 €/kg Phosphorus (2024 prices). Defra’s Enabling a Natural Capital Approach (ENCA) provide other options to value water quality improvement benefits (DEFRA, 2025). Based on the Farmscoper tool, the annual value of reducing a kg of nitrate/phosphorus in water is estimated at about 1.17 and 39.76 £ (2021 prices). These values are based on estimation of economic damages of nitrate and phosphorus on a range of ecosystem services, including drinking water quality, fishing, bathing water quality and eutrophication. Therefore, in the case of the Forth catchment, the replacement cost approach seems to overestimates the societal benefits of nitrogen removal but generates estimates in line with the Farmscoper tool for phosphorus retention. Note that both the replacement costs based and damage costs based values assume that all units of N retention provide the same benefit, i.e. each kg of N reduction has the same value. In reality, N emission reductions in sub-catchments where water quality is high have a lower value than N emission reductions in sub-catchments where water quality is poor. For instance, in the Forth Catchment restoration scenario, the nutrients retention benefit provided by peat restoration areas located in the Eastern part of the catchment, where water quality is mostly “moderate” to “good”, is likely higher than the benefit provided by peat restoration areas located in the North-Eastern part of the catchment, where water quality already reaches a “good” to “high” status. Figure 8 – Nutrients retention benefit in the Forth Catchment. Finally, both replacement costs based and damage costs based values integrate the value of multiple ESs related to nutrients content in surface water. Therefore, there is a risk of double-counting benefits if waterrelated ES benefits are also taken into account when valuing nature recreation (Section 10) or habitat provision benefits (Section 11). Nutrient retention MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 38 Lessons learned & recommendations Ø Use robust models like SWAT+ for accurate nutrient retention modelling, especially when restoration impacts are complex; simpler tools (e.g. InVEST, ESTIMAP) lack details on key processes. The MERLIN workflow provides a good starting point. Ø Use national data for damage costs when available (e.g. UK, NL, Germany) for more context-specific and robust valuation in CBA. Ø Avoid double counting, especially when other methods are included in the cost-benefit analysis (SP; WTP). Ø When using a replacement cost method, ensure there is societal demand for the replacement and that replacement is the least-cost alternative. Further reading: Ø Seitzinger, S., Harrison, J. A., Böhlke, J. K., Bouwman, A. F., Lowrance, R., Peterson, B., Tobias, C., & Drecht, G. V. (2006). Denitrification Across Landscapes and Waterscapes: A Synthesis. Ecological Applications, 16(6), 2064–2090. https://doi.org/10.1890/1051-0761(2006)016[2064:DALAWA]2.0.CO;2 Ø Van Grinsven, H. J. M., Holland, M., Jacobsen, B. H., Klimont, Z., Sutton, M. a., & Jaap Willems, W. (2013). Costs and Benefits of Nitrogen for Europe and Implications for Mitigation. Environmental Science & Technology, 47(8), 3571–3579. https://doi.org/10.1021/es303804g Ø Gourevitch, J. D., Koliba, C., Rizzo, D. M., Zia, A., & Ricketts, T. H. (2021). Quantifying the social benefits and costs of reducing phosphorus pollution under climate change. Journal of Environmental Management, 293, 112838. https://doi.org/10.1016/j.jenvman.2021.112838 Drought mitigation MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 39 9 Drought mitigation 9.1 ES logic chain Peatland and wetland restoration increases the sponge function of a landscape: the soil moisture content (and water available to plants) is higher, and excess water is absorbed during floods and slowly released to ground – and surface water during dry periods. This process supports higher baseflows in rivers and improves the recharge rates of both natural and artificial reservoirs, including aquifers. As a result, water availability during drought conditions is improved. This benefits 1) Hydropower generation and 2) Irrigated agriculture, cooling (e.g. byers et al) and drinking water supply, which depend on stable water availability and increased base flows. In large rivers and smaller streams, interventions such as re-meandering and other measures which reduce flow velocity help reduce riverbed incision. Riverbed incision exacerbates the disconnection of rivers from their floodplains and lowers groundwater tables in the vicinity of the river under low flow conditions. By maintaining higher water tables during dry periods (alongside higher baseflow), these measures contribute to 3) Inland shipping, by preserving navigable water levels; and 4) Water-based recreation, by sustaining water quality and water levels. Table 11 – Drought mitigation logic chain. Ecosystem types Factors determining supply Factors determining use Potential physical metric(s) Benefits Main users and beneficiaries Ecological Societal Peatlands, wetlands, rivers and lakes Groundwater recharge (infiltration) and water storage in landscape Water infrastructure – built to drain or to retain Demand & infrastructure for irrigation water, Average water yield in dry months Increased agricultural yield Farmers Peatlands, wetlands, rivers and lakes Groundwater recharge (infiltration) and water storage in landscape Water infrastructure – built to drain or to retain Presence of infrastructure to retain & release water Outflow of reservoirs and lakes Energy production Energy companies Large rivers Riverbed elevation/ water depth in dry season; bed sedimentation /erosion Water level regulation infrastructure (weirs, locks, groynes) Depth of ships, load capacity Water depth during low-flow Navigation cost (function of load capacity) Navigation/ transport sector Large rivers & streams Riverbed elevation/ water depth, water quality, upstream landscape retention capacity Water level regulation infrastructure (weirs, locks, groynes) Infrastructure for water-based recreation (marina's, rental companies, beach) Changes in water level Water-based recreation water-based recreation businesses 9.2 ES quantification Drought mitigation refers to the capacity of freshwater ecosystems—such as wetlands, floodplains, and riparian zones—to store, retain, and slowly release water. Hydrological models can estimate how much water is stored in floodplains or wetlands and how this affects downstream availability during droughts and erosion/sedimentation processes. Exactly what is to be modelled depends on the goal, but could include modelling of water retention and release at the basin scale to estimate probability of flow rates (e.g. with SWAT+, Garcia et al. (2025), water levels under low discharge at the river stretch level, implications for the fill level of natural or man-made reservoirs (under conditions of reservoir operation) and/ or groundwater modelling. Drought mitigation MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 40 9.3 ES valuation: Water availability for crop production After modelling expected impacts of interventions on baseflow and/ or natural/artificial reservoirs, the implications of changes in water availability on crop production can be monetised. There are various approaches to calculate the economic value of irrigation water: Ø Using a production function to either estimate the additional yield generated by an additional unit of water (marginal value product) or the residual value of water after accounting for other inputs (residual value) (Steduto et al., 2012; Hack-ten Broeke et al., 2019; Bashe et al., 2022). Ø Market price: actual price of irrigation water (m3) in places where such a market exists, e.g. through an irrigation scheme, (Karabulut et al., 2016). Ø Revealed/ stated preference: the value of irrigation water can be embedded in land rental prices, which can be derived via hedonic pricing (Bashe et al., 2022): The price of agricultural land is determined by predicted profitability. Availability (and reliability) of irrigation water is one of many factors that influence the rent price. Another approach is to derive the willingness to pay via contingent valuation (large surveys asking farmers how much they are willing to pay for irrigation water). Ø Replacement costs: by calculating the alternative price for (constructed) water storage infrastructure (Esen et al., 2023)-. 9.4 ES valuation: Hydropower - sediment control and base flow Hydropower generation is increasingly vulnerable to the impacts of climate change. Hydropower facilities are designed based on historical flow regimes, which in turn depend on long-term precipitation patterns and the regulating ecosystem services provided by upstream watersheds (Spolum, 2021). As these patterns shift, the reliability and efficiency of hydropower generation are at risk. Hydropower systems generally benefit from stable, predictable flows throughout the year, and limited reservoir sedimentation. Sedimentation reduces the storage capacity or the reservoir and is one of the driving factors in lifecycle cost. Reduced storage capacity makes the hydropower dam more reliant on seasonal flows, making it even more important that these are stable and predictable over the year. Peatland rewetting, upstream forest restoration and other restoration measures that improve upstream retention and slower discharge (e.g. remaindering of streams) help smooth flows over the year (Cities4Forests) and reduce sediment flows which drive costs for hydropower generation (Guo et al., 2023). Monetising those benefits requires economic modelling to understand how 1) changes in availability of water over the year relate to local energy market demands (which affect the price) and 2) how sedimentation volumes affect the lifespan and/ or lifecycle costs of the hydropower installation. The monetary value of sediment control can also be derived using cost-based methods, for example by calculating costs related to removing sediment in reservoirs (Esen et al., 2023). Box 9.1: : Impact of Groundwater Changes on Agricultural Revenue near the Rhine Branches The river Rhine in the Netherlands faces a disrupted sediment balance, which in turn causes riverbed incision and reduced lateral connectivity. This affects groundwater levels within and beyond the floodplains, which increases drought damage to (non-irrigated) crops. Kok et al. (2025) estimated impacts of floodplain reconnection and riverbed sediment dynamics on mean low groundwater tables (MLGT)—influenced by river stage-discharge relations, using the Dutch WaterWijzerLandbouw model (Hack-ten Broeke et al., 2019). This model is specifically designed to quantify impacts of hydrological conditions on yield, and provides spatially explicit estimates of potential gross revenue per plot (€) and expected yield loss (%) due to dry conditions. Under the baseline, losses due to low groundwater tables amount to € 86 million per year, in the area in which groundwater tables are influenced by the water table in the river (in places up to 10 km beyond the riverbed). Ultimately, the impact of changes in river water tables on groundwater-related yield loss was quite limited in the case study area (<1%). It should be noted that this study only reviewed the impact of changes in water tables on dry conditions: further dynamic groundwater modelling is recommended to assess the role of high groundwater tables, which may offset the limited benefits of MLGT increases (Bartholomeus et al., 2011). Drought mitigation MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 41 Though research on the impact of ecosystem restoration on hydropower generation is limited, existing literature shows promising results. For example, ecological restoration (including revegetation) in the Yellow River Basin in China - characterised by high sediment loads - enhances hydropower potential – reducing upstream sedimentation. There is a trade-off: restoring vegetation in the area reduced runoff rates throughout the year, but the benefits of reducing the lifespan of the installation by reducing sediment rates outweighed this effect (Wu et al., 2025). In California’s Sierra Nevada, forest restoration after wildfires has been shown to influence both the quantity and timing of water flows, benefiting downstream hydropower generation (Guo et al., 2023). When increased runoff is expected under climate change, as in Sweden’s Ume river, ensuring sustainable environmental flow can be achieved with minimal loss to hydropower production (Widén et al., 2024). 9.5 ES valuation: Inland shipping Inland waterways benefit from sufficient channel depth (or draught) during low-flow periods – this directly affects load capacity, and in turn transport costs for shipping. Freshwater restoration measures that reduce flow velocity and thereby riverbed erosion, and measures that increase the base flow can support higher channel depth during low flow. To evaluate the impact of impact of freshwater restoration measures on navigation requires 1) modelling of hydro-morphological changes over time (e.g. Cron et al., 2015) and 2) modelling of the impacts on shipping. The latter is typically done using a production function, calculating the impact of reduced draught on the carrying capacity on ships, as illustrated in Box 9.3. Box 9.2: : From reservoir fill level to (prevented) losses in irrigated agriculture (Sorraia, Rhine) Upstream land use changes can affect the water balance in a reservoir, and in turn the fill level of reservoirs used for irrigation and other water uses. In the MERLIN case on the Portuguese Sorraia basin, upstream (re)forestation scenarios are studied, comparing (non-native) eucalypt plantations and native oak forest restoration: the fill level of the reservoir decreases when eucalypt plantations are installed (modelled using SWAT2012). The impact on the agricultural productivity in the irrigated system was estimated by calculating Agricultual yields in the area combining crop-specific standard output coefficients (€/ha) for Portugal (building on EU-wide dataset from ESTAT 2024), which are farm-gate value of output (or gross revenue). Assuming in an irrigated system, the availability of water directly affects crop production (1:1.3, accounting for distribution losses), a 5% reduction in water inflow translated in a loss of €1.2 mln (or €21 million present value, at discount rate 3%). In the case of the Rhine, interventions to restore the sediment level in the riverbed altered the distribution of water across the river branches, benefiting the fill level of the IJsselmeer reservoir at the outlet of the IJssel branch. The economic impacts was calculated based on a price per m3, derived using a marginal product value approach: €0.55/m3 (including VAT, 2020), for the affected area (the national average lies at €0,67). In a situation where flow distribution is redirected towards the IJsselmeer, this gives an annual benefit of €7 mln (or €231 mln present value, at discount rate 2.25% up till 2120) Habitat provision MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 48 Box 11.1 : Habitat indicators in river-floodplain policy evaluation in Germany In Germany, several indicators have been applied in evaluating river management policy. In the RESI project, Podschun et al. (2018) assess and quantify a ESS habitat index for the habitats of floodplains and rivers separately, on a 0 to 5 point-rating scale. For floodplains, the rating is based on the existing biotope types and their quality and on an assessment at the river-floodplain section level. For the river itself, the rating is based on biologically relevant watercourse structures, using the Valmorph method (Roesch et al., 2025, based on Quick et al., 2020) to assess morphological quality and biological indicators from the WFD, including fish, benthic invertebrates and macrophytes. In the AMBERS project a similar approach is used, awarding a bonus or malus to the rating for changes in flooding regime and chemical status of the waterbody (Roesch et al., 2025). Figure 9 shows an application of the RESI index on the case of the Ammer (Bavaria). The index is based on biotope types, with value adjusted for land use intensity, floodplain connectivity, and conservation status (e.g., Natura 2000 habitats). Higher scores are assigned to areas with natural vegetation, low disturbance, and ecological uniqueness. The habitat quality increases in the riparian reforestation scenario because land use intensity is reduced and additional areas are allowed to undergo natural succession into riparian forest. These forests are rare and ecologically valuable in Germany, and therefore receive a higher score. A) Land cover in status quo B) Habitat provision index under status quo C) Land cover under 'riparian reforestation' D) Habitat provision index under reforestation Figure 9 – Restoration scenarios in comparison with status quo and its impact on changing habitat provision in Germany (Becker et al., 2022) Habitat provision MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 49 11.3 ES valuation Beyond quantitative metrics to assess the impact of FWE restoration on habitat provision, various methods can be used to value this service. Valuation approaches for habitat provision are typically non-market-based, involving proxy methods such using stated preference to derive a willingness to pay (WTP), or restoration and compensation costs. Stated preferences approaches are less accepted in some countries, which instead suggest presenting a habitat indicator alongside the cost-benefit analysis (e.g. in Dutch CBA guidance, Renes & Romijn, 2013) or using a restoration/ compensation cost approach. 11.3.1 Stated preferences Stated preference methods, such as contingent valuation and choice experiments, can be used to monetize the habitat provision service by capturing the non-use values people assign to biodiversity and ecosystem restoration. These values reflect the societal importance of preserving nature for future generations (bequest value) and the intrinsic worth of ecosystems regardless of direct use (existence value). These methods estimate the willingness to pay (WTP) for improvements in wildlife habitats, which, which careful design, can be used to distinguish the use value (e.g. recreation) from non-use values (reflecting bequest and existence values). This is important to prevent double counting with other ecosystem services such as recreation (see Box 11.2 : Habitat provision - policy achievement indicator for the Rhine in the Netherlands (MERLIN CS 04) As part of a wider ecosystem service assessment of river-floodplain management scenarios of the Rhine in the Netherlands, Kok et al. (2025) developed a habitat provision indicator based on policy achievement, in order to assess the effectiveness of river-floodplain management (RFM) strategies in meeting habitat targets set under the Programmatic Approach Large Waters (PAGW). Habitat targets under the PAGW were determined using the LARCH model (van der Sluis et al., 2020), which uses species-specific area requirements to derive minimum habitat sizes and connectivity thresholds needed to support sustainable populations of key target species. The policy achievement indicator used in Kok et al. (2025) s a similarity index ranging from 0 to >1, comparing the actual surface area of natural ecotopes under each RFM strategy to the desired ecotope distribution stipulated in the PAGW. Box 11.3 : National habitat provision indicator for scenario analysis in Germany and the Netherlands (MERLIN CS 04) The MAES framework supports biodiversity indicators that can be used in national biodiversity monitoring and planning (Maes et al., 2016); for example the Biodiversity Intactness Index (BII; an aggregated indicator of species abundance and reference conditions) and Mean Species Abundance (MSA). Additionally, several member states have developed their own indicators. In Germany the ecosystem service indicator “habitat provision” can be assessed, calculated and monetised based on the value of a biotope´s condition “Biotopwertpunkte” or Biotope value system. This value type describes the change in condition of a biotope before and after a measure. The condition type is regulated by the “Federal Compensation Ordinance” or in some cases the state´s individual compensation ordinance regulations, which use biotope value points to determine ecological compensation measures for interventions in nature (Ekinci et al., 2022). For the ordinance, calculating the condition of an ecosystem per hectare, based on criteria including age of an ecosystem, occurrence of threatened species, and naturalness. Classification of ecosystem condition in ratios from 0 (paved ground) to 24 (intact peat bogs, old (semi-) natural forests). Attribution of value to a comprehensive and consistent system of nation-wide data on ecosystem extent and condition based on German land cover, forestry, an EU directive´s data. Data weighing and aggregating to a physical number (e.g. C02). In the Netherlands, the biodiversity points approach ('natuurpunten') (Bos and Ruijs, 2019), a standardised, transparent method to quantify the ecological impact of project alternatives, is recommended for use in CBA. The approach is a threat-weighted Ecological Quality Area, and is calculated based on the size of the ecosystem affected, the ecological quality (a score or 0-1) based on species presence v.s. intact reference ecosystems and a threat weight – reflecting the rarity and conservation priority of the ecosystem. A noted limitation is that the approach is less developed for water-quality related biodiversity due to diffuse impact areas and context-dependent quality metrics. Habitat provision MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 50 also chapter 12). If a direct valuation study is out of scope – they are resource-intensive – a meta-regression analysis can be used to derive a benefit transfer function. For example, Brouwer & Sheremet (2017) explicitly study the WTP for the function 'wildlife habitat' as a result of river restoration. Box 11.4 on the Rhine case provides an example of the application of such a benefit transfer function in a cost-benefit analysis. 11.3.2 Restoration and Compensation Cost Approaches Ecological damage and/ or the value of ecological restoration can be monetized using a cost-based approach, under the assumption that restoration or exchange yields an equivalent amount of ecosystem services. This cost-based value reflects total economic value, including use and non-use components, and can therefore not be used alongside other ecosystem service valuations to prevent double-counting. There is no standardized or widely accepted approach for monetizing losses or gains in habitat provision: existing methods vary significantly across regions and institutions. The following examples from the EU, Germany, and the USA illustrate how the cost-based approach to value habitat provision can be applied in practice. The EU Handbook on the external costs of transport (van Essen et al., 2019) includes a method for estimating the negative effects on nature and landscape – including effects of habitat loss and degradation. Based on restoration costs, cost factors for all EU countries have been calculated, ranging from €3000-11.000 €/km/year for 'damage to inland waterways'. In Germany, a standardized biotope value point system is used to guide restoration investment and compensation requirements under the planning law (Schweppe-Kraft et al., 2020). Habitats are scored from 0 (sealed surfaces) to 24 (high-value habitats like intact peat bogs) using a standardized ecological matrix, with adjustments of ±3 points allowed based on field condition (e.g. degradation or exceptional quality). Three methods can be used to determine the monetary value per biotope value points (Ekinci et al., 2022; SchweppeKraft et al., 2023): 1) restoration investment cost (based on cost for achieving EU habitats directive: 3,634€ per biotope value point), 2) compensation market value (preliminary findings from habitat banking transactions; 16,000€ per point) and 3) WTP per point (based on WTP for national nature conservation programmes: 7800€). In the USA, a similar approach is used to determine required compensation or restoration investment in case of ecological damage, which can also be used to assess the value of pro-active restoration projects: the Habitat Equivalency Analysis (HEA) (NOAA, 2006). The HEA uses a service-to-service approach (assuming a one-to-one trade-off in ecological services), but can also incorporate stated preference methods, e.g. to estimate non-use values, or address equity and distributional concerns. In the service-to-service approach, restoration is scaled so that quantity and quality of ecosystem services provided over time by the restoration project match those lost by the damage. The approach includes 1) Estimating the loss of ecological services over time until natural recovery. 2) Identifying restoration alternatives: These include both primary restoration (returning the resource to baseline) and compensatory restoration (offsetting interim losses), and 3) Scaling restoration: Determining the size and scope of restoration needed so that the present discounted value of ecological gains equals the present discounted value of losses. The approach indirectly accounts for restoration time: the longer it takes for a system to recover, the greater the interim loss. Box 11.4 : Benefit transfer for valuation of habitat provision in MERLIN CS04 Room for the Rhine branches In the Rhine Branches study (Kok et al, submitted), a benefit transfer function was developed to estimates households’ willingness to pay (WTP) for biodiversity improvements based on a meta-regression analysis of European wetland valuation studies (Grossmann, 2012a). The function incorporates key variables such as the size of the restored area, average household income, and the degree of ecological improvement (linked to the policy achievement indicator, see box 11.2). The resulting WTP estimates ranged from €5.50 to €28.70 per household per year, depending on the degree of restoration in the strategy. These estimates were aggregated across the population to using a uniform distance decay function - assuming that the value people assign to habitat improvements diminishes with increasing distance from the restoration site due to place attachment. The market area was defined as 35 km radius around study area, based on empirical results for water quality improvements in the Netherlands (Schaafsma et al., 2012). Habitat provision MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 51 Figure 10 – Biotope overview map of Germany and assigned biotope value points per biotope type (Ekinci et al., 2022). Lessons learned & recommendations Ø Habitat provision is an intermediate ESS that supports other services by maintaining ecological structures and biodiversity; however, the bequest or existence value of this service can be monetized using stated preference methods or benefit transfer functions. There is no standardized method for this. Primary valuation studies can be resource-intensive, so benefit-transfer of WTP may be more realistic. Ø Quantitative indicators like biotope value points can be used to reflect (changes in) habitat quality (or quantity), and can be either 1) presented alongside monetized benefits in a CBA, 2) monetized, as demonstrated by Schweppe-Kraft et al. (2023) or 3) embedded in a benefit transfer function, as illustrated in the Rhine case (Box 11.4). Monetary valuation of biotope value points (e.g. using a restoration costs approach) is a relatively new field and deriving them can be methodologically challenging. Ø Cost-based methods can be used to represent the total economic value for ecosystem restoration – as these estimates cover all ecosystem services they can therefore not be used alongside other ecosystem services. Habitat provision MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 52 Further reading: (Schweppe-Kraft et al., 2023) Ecosystem services for conservation of biodiversity/species and habitats. In: Nature under pressure. (Brouwer and Sheremet, 2017) The economic value of river restoration – meta-analysis, including assessment of WTP for habitat provision. Biomass provision MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 53 12 Biomass provision 12.1 ES logic chain The biomass provision ES is the contribution of ecosystems to the growth of biomass that is harvested by economic agents for various uses such as production of food, fibre, fodder and energy. Table 14 – Biomass provision logic chain. Ecosystem types Factors determining supply Factors determining use Potential physical metric(s) Benefits Main users and beneficiaries Ecological Societal Peatlands, wetlands, floodplains Soil fertility, climate, water supply Farm management Demand for biomass Tons dry matter per ha Food, fibre, fodder, energy Farmers, households, Energy, Manufacturing Freshwater ecosystems restoration may have multiple effects on biomass provision. The effects of FWE restoration may involve changes in land use. Such changes may lead to foregone biomass provision, for instance when a floodplain farmland is converted to natural ecosystems. By convention, foregone biomass provision is accounted for as costs in a CBA (opportunity costs). Land use changes may also provide opportunities for new production, such as pluviculture on wetlands (de Jongh et al., 2021; Wichmann, 2017) or extensive livestock grazing in floodplains, accounted for as biomass provision benefits. FWE restoration may also bear indirect effects on biomass provision, through changes in hydrology (e.g. groundwater recharge) or habitat for functional biodiversity (pollinators, natural pest enemies). Such indirect effects may be estimated by quantifying changes in intermediate ES supporting biomass provision (e.g. water provision, pollination, natural pest control). This section is focused on estimating the direct effects of restoration on biomass provision. Effects of hydrology changes on agricultural production are addressed in Section 8 on drought risk mitigation. 12.2 ES quantification Biomass provision is straightforwardly quantified as harvested biomass, e.g. tons of reed dry matter per ha of peatland, using existing national agricultural and forestry statistics or field survey data. In the absence of national data, EU-wide datasets on crop types areas and yields also exist (see Annex 1). As harvested yields can vary greatly from year to year due to weather fluctuations, it is advisable to use multiple year averages. Note that in most cases, biomass provision is a joint production process that involves human interventions (management, harvest) in combination with ecosystems contributions (e.g. water and nutrients cycling, pollination and pest control). Methods are being developed to disentangle human and ecosystem contributions (Wijngaart et al., 2019). In the context of a CBA of freshwater ecosystems restoration, separating human from ecosystem contributions in biophysical terms is not relevant, and the service can be quantified by harvested yields. 12.3 ES valuation Monetary valuation studies of biomass provision ES usually attempt to value specifically the contribution of ecosystems to biomass provision. Three methods can be applied: production functions, land rental prices and resource rents. The contribution of land to biomass provision can be derived from a production function in which the production output is a function of land (the ecosystem), labour, capital and other factors. Such production functions allow to assess the marginal productivity of the land (ecosystem) to output, and, multiplying this marginal land productivity by the output price, the biomass provision ES value. In practice, applying this method requires the availability of farm level data on farm inputs and outputs which is often not available. As workaround, land rental prices can be used to approximate the marginal land productivity. However, land rental prices may not always represent a valid approximation of marginal land productivity, for instance in case of market distortion or land speculation. In the resource rent method, the contribution of land to biomass provision is the residual value of the biomass output after subtracting all other inputs values (farm inputs, capital, labour), derived from national statistics on farm incomes and production costs. A drawback of the resource rent method is that provides very variable estimates, sometimes negative, due to sometimes large annual variations of output yields and prices. It is therefore advised to use averages over multiple years. In a social CBA, separating the values of ecosystem and human contributions is not relevant. It is more appropriate to use the total value of biomass production, i.e. total yield multiplied by prices, which indicates the benefit for societal welfare of the restoration project. However, the resource rent method, which is an Biomass provision MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 54 estimation of farming net incomes (after deducting production costs from product sales), can be useful to determine the effects of restoration on the farming sector (see Section 13). Lessons learned & recommendations Ø For a social CBA, it is recommended to use biomass yields and market prices to value the biomass provision ES. Ø Methods used in the ES science domain to separate ecosystem and human contributions to biomass provision are not adapted to social CBA but may be useful to assess the effects of restoration on the farming sector. Further reading: Ø Joosten et al. (2016) Paludiculture: sustainable productive use of wet and rewetted peatlands (Vol. 10). Connecting a social CBA to private financing of freshwater ecosystems restoration MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 55 13 Connecting a social CBA to private financing of freshwater ecosystems restoration This report provides guidance and recommendations for carrying out a social Cost - Benefit Analysis of FWE restoration. A social CBA evaluates costs and benefits of restoration for the whole society, aggregating benefits and costs of all public and private stakeholders. The primary aim is to determine whether it is in the interest of society as a whole to invest in the restoration project, not to estimate costs and benefits of FWE restoration for specific economic sectors. However, through the identification and quantification of a large range of societal benefits, a social CBA may be a valuable starting point for developing private (co)funding sources, e.g. through conventional markets (e.g. valorisation of biomass, recreation services), payment for ecosystem services or environmental markets (e.g. carbon credits) (see MERLIN Deliverable 3.6). This report presents several methods for the monetary valuation of ESs in a social CBA context. Not all these methods deliver useful outputs for informing private funding opportunities. To be useful in development of a business case, special attention should be paid to attributing costs and benefits to specific (private, local) stakeholders, and valuation methods should be related to revenues or costs incurred by specific sectors; there is a disconnect between the sometimes abstract values of ecosystem values and practical application in nonpublic settings (Siebers et al, 2025). Table 15 presents an overview of which monetary valuation methods are preferred from a welfare-economics perspective, and which methods are most suited for informing a (co)funding/ financing strategy. For instance, while valuing carbon sequestration with social costs of carbon or efficient prices is recommended in a welfare economics perspective, in a context of generating revenues only the actual carbon credit prices on voluntary carbon market are relevant. Table 15 – Recommended monetary valuation approaches for social vs business case. Ecosystem Service Social CBA Business case Potential funding source Flood risk mitigation Avoided damage costs Replacement costs Avoided damage costs Businesses (farms, industries, commercial units) exposed to flood risks and flood insurance sector Climate change mitigation Social cost of carbon, efficient carbon prices Carbon market prices (voluntary or compliance) Businesses compensating their GHG emissions Nutrients retention Replacement costs Avoided damage costs/social costs of nutrients Water treatment costs Loss of recreation value Water supply companies Water-based recreation businesses Drought mitigation Market price/ production function Replacement cost Avoided damage costs, Stated preference Avoided damage costs Market price/ production function Navigation sector, hydropower sector, water-based recreation businesses, water users (drinking water, irrigation, cooling water) Recreation Contingent valuation/Choice Experiment Nature recreation business activity (e.g. resource rent of recreation sector) Water-based recreation businesses Habitat provision Contingent valuation/Choice Experiment Restoration costs Habitat banking, biodiversity market prices Businesses compensating their impacts on biodiversity Biomass provision Market prices, resource rent, land prices, production function Resource rent, production function Farmers, land managers Other key difference between social CBA and business cases include the timing of benefits, the time horizon of the assessment and the degree of details in the cost assessment. For instance, in a social CBA, carbon Connecting a social CBA to private financing of freshwater ecosystems restoration MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 56 sequestration benefits are accounted for from the first year when restoration effects on carbon sequestration take place. In a business case, the benefits only start to accrue when carbon credits are sold (depending on the contract, this may be a one-time payment or periodic), which may be at least 3 to 5 years after the first year when restoration effects on carbon sequestration take place. This can have important impacts on the financial viability of a project, especially since the discount rate used in private business plans is higher than in a social CBA. Similarly, the time horizons of interest in private sector are much shorter (2-10 years) than the horizon used to assess the societal benefits in a social CBA (20-100 years, see section 2.3). To inform a (co-)funding or financing strategy, a detailed understanding of the project cost structure and timing across the project lifecycle is needed, whereas the assessment of costs in a social CBA relies on unit cost estimates (see also section 3). Though a social CBA is not directly transferable to a private investment setting, it provides valuable inputs for developing private (co)funding strategies. Estimates of the biophysical effects of restoration - such as tons of C sequestered, reduction in flood risk or maintenance of base streamflow – can deliver key inputs to assess benefits and costs of restoration for economic sectors. When doing so, specific monetary valuation methods that provide usable monetary estimates for private stakeholders should be used – those that reflect market prices, cost savings or revenue generation potential. References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 57 References Aerts, J.C.J.H., 2018. A review of cost estimates for flood adaptation. Water (Switzerland). https://doi.org/10.3390/w10111646 Aerts, J. C. J. H., Bates, P. D., Botzen, W. J. W., Bruijn, J. D., Hall, J. W., Hurk, B. V. D., Kreibich, H., Merz, B., Muis, S., Mysiak, J., Tate, E., & Berkhout, F., 2024. Exploring the limits and gaps of flood adaptation, Nature Water, 2(8): 719–728. https://doi.org/10.1038/s44221-024-00274-x Aalbers R., Renes, G., and Romijn, G. (2017). WLO-klimaatscenario’s en de waardering van CO2-uitstoot in MKBA's. CPB/PBL Achtergronddocument, Den Haag, 23 november 2016. Ayres, A., Gerdes, H., Goeller, B., Lago, M., Catalinas, M., García, Á., Brouwer, R., Sheremet, O., Vermaat, J., Angelopoulos, N., Cowx, I., 2014. Inventory of river restoration measures: effects, costs and benefits, REFORM project, Berlin: Ecologic Institute, 95 pp. Bashe, T., Alamirew, T., Dejen, Z.A., 2022. Estimating the economic value and economic return of irrigation water as a sustainable water resource management mechanism. Sustainable Water Resource Management (10)42. https://doi.org/10.1007/s40899-022-00764-4 Birk, S., Weigelt, C., Borgwardt, F., Kail, J., 2025. Freshwater restoration effects on biodiversity and ecosystem services : a Delphi survey. Restor. Ecol. 1–13. https://doi.org/10.1111/rec.70119 Bleicher, K., Kunder, S., Fenske, M., Nilson, E. und Fuchs, E. (2024): AMBERS – Ansätze einer multikriteriellen Bewertung von Maßnahmen an Bundeswasserstraßen. Bundesanstalt für Gewässerkunde, Koblenz, BfG-Berichtsnummer NN DOI. NN Brander, L.M., Florax, R.J.G.M., Vermaat, J.E., 2006. The empirics of wetland valuation: A comprehensive summary and a meta-analysis of the literature. Environmental and Resource Economics (33): 223-250. https://doi.org/10.1007/s10640-0053104-4 Brink, C., van Grinsven, H., Jacobsen, B.H., Rabl, A., Gren, I.-M., Holland, M., Klimont, Z., Hicks, K., Brouwer, R., Dickens, R., Willems, J., Termansen, M., Velthof, G., Alkemade, R., van Oorschot, M., Webb, J., 2011. Costs and benefits of nitrogen in the environment, in: Sutton, M.A., Howard, C.M., Erisman, J.W., Billen, G., Bleeker, A., Grennfelt, P., van Grinsven, H., Grizzetti, B. (Eds.), The European Nitrogen Assessment: Sources, Effects and Policy Perspectives. Cambridge University Press, Cambridge, pp. 513–540. https://doi.org/DOI: 10.1017/CBO9780511976988.025 Brouwer, R., Sheremet, O., 2017. The economic value of river restoration, Water Resources and Economics, 17: 1-8. https://doi.org/10.1016/j.wre.2017.02.005 Bos, F., Ruijs, A., 2019. CPB background document - Biodiversity in Dutch CBApractice. The Hague. https://www.cpb.nl/system/files/cpbmedia/omnidownload/CPB-Background-Document-feb2019-Biodiversity-in-the-Dutchpractice-of-cost-benefit-analysis.pdf Bünger, D.B., Matthey, D.A., 2018. Methodenkonvention 3.0 zur Ermittlung von Umweltkosten. Methodische Grundlagen, Dessau-Roßlau: Umweltbundesamt, 61 pp. Contor, H., 2025. Integrated Unpacking Methods for Integrated Assessments of Nature-Based Solutions Guideline for the European Investment Bank and public authorities IISD REPORT, 38 pp. Cools, J., Interwies, E., 2024. Cost-benefit assessment for nature-based solutions for flood mitigation. Developed under the Framework Contract ‘Water for the Green Deal’ - Implementation and development of the EU water and marine policies (09020200/2022/869340/SFRA/ENV.C.1), Luxembourg: Publications Office for the European Union, 21 pp. Cron, N., Quick, I., Vollmer, S., 2015. Quantitative Evaluation of Hydromorphological Changes in Navigable Waterways as Contribution to Sustainable Management, in: Hipel, K., W., Fang, L., Cullmann, J. & M. Bristow (eds.) Conflict Resolution in Water Resources and Environmental Management. Springer International Publishing, pp. 245–262. https://doi.org/10.1007/978-3-319-14215-9_14 De Blaeij, A., Linderhof, V., Polman, N., Reinhard, S., 2009. Optimizing Commercial Wetlands in Rural Landscapes Optimizing Commercial Wetlands in Rural Landscapes Optimizing Commercial Wetlands in Rural Landscapes, Paper presented at The International Conference on Landscape Economics. https://edepot.wur.nl/12701. DEFRA, 2025. Enabling a Natural Capital Approach. https://www.data.gov.uk/dataset/3930b9ca-26c3-489f-900f6b9eec2602c6/enabling-a-natural-capital-approach. de Grave, P (2021), Gebruikershandleiding OKADER 2021 - Opgave en Kosten Analyse Dijkversterking en Rivierverruiming. Deltares. de Jongh, L., de Jong, R., Schenau, S., van Berkel, J., Bogaart, P., Driessen, C., Horlings, E., Lof, M., Mosters, R., Hein, L., 2021. Natuurlijk Kapitaalrekeningen Nederland 2013-2018: rapport, Wageningen: Wageningen University & Research, 77 pp. https://edepot.wur.nl/566544 De Knegt, B., Biersteker, L., Van Eupen, M., Van Der Greft, J.G.M., Heidema, A.H., Koopman, R., Jochem, R., Lof, M.E., Mulder, H.M., Van Rijn, P., Roelofsen, H.D., De Vries, S., Woltjer, I., 2022. Natural Capital Model, WOt-technical report, Wageningen: WOT Natuur & Milieu, 236: 251 pp. De Nocker, L., Verachtert, E., Broekx, S., Poelmans, L., Brabers, L., Liekens, I., De Valck, J., Van Der Meulen, M., 2016. Kwantificering en waardering ecosysteemdienst Recreatie-methode 2016. Deltares, 2022. Delft3D FM Suite 2D3D, https://www.deltares.nl/en/software-and-data/products/delft3d-flexible-meshsuite (last accessed 14.08.23). Deltares, 2020. SOBEK 2 Suite, https://www.deltares.nl/en/software-and-data/products/sobek-suite (last accessed 14.08.23). References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 64 The plugin only uses a hydrological model, SWAT+, which allows to model the effect of restauration on the return periods/probabilities of peak stream flows and associated flood events (see MERLIN Deliverable 3.3). SWAT+ generates average daily flow rate output values at river section level, that can be used to estimate restauration induced changes in peak flow probabilities at river section level. The probability of flood damage in a raster cell of the flood damage map is then derived from the change in peak flow probabilities of the closest (Euclidean distance) river section. Therefore, in Equation 1, only Ri changes between the baseline and restoration scenarios. Di is kept constant and has to be provided as input data by users of the plugin. The next section provides guidance on methods and data sources to generate these inputs. Figure 12 – Estimation of restoration effects on floods EAD Methods and data sources to produce flood damage costs maps A flood damage costs map corresponding to each flood return period Ri is needed to calculate Di (see Eq. 1). Flood damage costs maps provide an estimate of flood damage costs for each map unit j (Dij), which are summed over the total number map units, c, to calculate Di. Flood damage costs for each map unit j are the product of the land use value of this map unit (Vjk) and the damage fraction (DFijk) due to floods (Eq. 2). To obtain Vjk a land use map and economic values of each land use class of the land use map are needed. Damage fraction (DFijk) due to floods in map unit j are derived from flood depth – damage functions, that relate DFijk to flood depth in map unit j, Depthij. Flood depth in map unit j is obtained from a flood extent and depth map. Thus, four different data sources are needed: a land use map, land use classes economic values, flood depth – damage functions and flood extent and depth maps for each flood return period. The Freshwater Information System for Europe - Flood Risk Areas Viewer of the EEA Europe (https://discomap.eea.europa.eu/floodsviewer/) provides links to flood extent and depth maps produced by EU member states in response to the EU Floods Directive. National data sources on land use maps, land use classes economic values and flood depth – damage functions are the preferred option and may be available in many EU countries. When it is not the case, default EU data may be used. We present hereunder an approach based on EU data that has been applied to MERLIN CS CBAs. The global database in Huizinga et al. (2017) provides Europe-specific flood depth – damage functions (damage % as a function of flood depth) for 6 land use classes (residential, commercial, industry, transport, infrastructure, agriculture). Moreover, this dataset provides country specific maximum damage values for those 6 land use classes. Flood damage (€) Flood probability (%)/ return period 0,1 % - 1/1000 years 1 % - 1/100 years 10 % - 1/10 years D1/1000 D1/100 D1/10 Restauration eFect Baseline Ri Restauration Ri EADrestauration EADbaseline References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 65 Land use data was obtained from the Corine Land Cover dataset (https://land.copernicus.eu/en/products/corine-land-cover). CLC classes were cross-linked to the 6 land use classes in Huizinga et al. (2017) (Table 16). Table 16– Cross link of CLC classes and Huizinga et al. (2017) land use classes CLC class level 1/2/3 Huizinga et al. (2017) land use class 1.1 Urban fabric (including all level 3 sub-classes) Residential 1.2.1 Industrial or commercial units Commercial & Industry 1.2.2 Road and rail networks and associated land Infrastructure 1.2.3 Port areas Transport 1.2.4 Airports Transport 1.3.3 Construction sites Residential 1.4.1 Green urban areas Residential 1.4.2 Sport and leisure facilities Residential 2. Agricultural areas Agriculture Note: Level 3 classes mineral extraction sites (1.3.1) and dump sites (1.3.2), as well as level 1 classes (including all level 2 and 3 sub-classes) forest and semi-natural areas, wetlands and water bodies did not have any corresponding land use class in Huizinga et al. (2017). Therefore, no flood damage is calculated on these land use/land cover classes. Flood depth – damage functions in Huizinga et al. (2017) consist of look-up tables with 9 discrete flood depth points (0, 0.5, 1, 1.5, 2, 3, 4, 5, 6 meter) and their corresponding damage fractions (range 0 – 1). Using these values, we fitted quadratic/cubic functions for each land use class (Table 17). Table 17– Flood depth – damage functions Land Use Class Depth Damage Function (y = Damage Fraction (0-1), x = flood depth in meters) Residential y = -0.0275x2 + 0.3166x + 0.066 Commercial y = -0.0284x2 + 0.3398x - 0.0069 Industrial y = -0.0225x2 + 0.3055x - 0.004 Transport y = -0.054x2 + 0.4659x + 0.0803 Infrastructure y = -0.034x2 + 0.3579x + 0.0576 Agriculture y = 0.0097x3 - 0.1288x2 + 0.5882x + 0.0273 Commercial & industrial* y = -0.0255x2 + 0.32265x - 0.00545 * As the land use class Commercial and Industrial correspond to only one class in the CLC dataset, we estimated a joint depth – damage function for the Industrial or commercial units CLC class by averaging the coefficients of the commercial and industrial depth – damage functions. These functions allow to determine a damage fraction for flood depth values (between 0 and 6 meter) that are provided by flood extent and depth maps. When flood and extent maps provide single depth values per map unit (continuous data), those flood depth values can be directly used in the depth damage functions. When References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 66 flood and extent depth maps provide ranges for depth valuers per map unit (categorical data), a unique flood depth value per depth category needs to be determined. For instance, the flood extent and depth maps used in the Forth CS had three flood depth classes: Less than 0.3m, 0.3m to 1m and Greater than 1m. To estimate damage fractions, we used the mid-range value for the two first classes (0.15 and 0.65 m.) and a value just above the lower bound for the third class (1.1 m.) to avoid over-estimation. Finally, the economic value of land use classes needs to be converted from 2010 euros to current year euros to consider inflation. 𝑉52 =%𝑉#6$6 %×%(1+𝑟)^(528#6$6), Eq. 5 Where Vyr is the land use value in current year euros, r is the average inflation rate between 2010 and current year, yr is the current year. The average value of the commercial and industrial classes is assigned to the Industrial or commercial units CLC class. References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 67 Annex 3: Detailed description of MERLIN plugin to value nutrient retention benefits The MERLIN plugin values the nutrient retention service provided by ecosystems in a water catchment on the assumption that the service would be replaced by constructed wetlands if it did not exist. This approach was adapted from La Notte et al. (2017). Replacement constructed wetlands are dimensioned to provide a nutrient retention service equivalent to the service provided by the restoration of ecosystems. The biophysical outputs of SWAT+ (see D3.3) - water flows, nutrients inputs and losses – are used to quantify the provided nutrient retention service. The service can be modelled for Nitrogen, Phosphorus or both Nitrogen and Phosphorus retention. In the equations below, N means Nutrients, either Nitrogen or Phosphorus. The replacement CW surface area is calculated for each landscape unit and the outlet channel in the water catchment using Eq. 1. 𝐴/=%$#.; *%∙ln(,!8+,∗ ,%8+,∗*, Eq. 1 Where As is the CW surface area (in m2), Q is the monthly streamflow (in m3/month), k is an areal constant (m/year); for nitrogen removal 𝑘 = 𝑘#6.𝜃(<8#6) where T is temperature of the water in degree Celsius, k20 is 30.6 and θ is 1.102 for FWS CW (Kadlec & Wallace, 2009; La Notte et al., 2017), ci is the outlet N concentration (in mg/L) in the baseline scenario, ce is the outlet N concentration (in mg/L) in the restoration scenario, c* is the background N concentration, assumed to be 0 mg/L. The ratio ci/ce, i.e. the concentration of N outputs in the baseline scenario over the concentration of N outputs in the restoration scenario, quantifies the nutrient retention provided by restoration in each landscape unit and the outlet channel, which is also dependent on the volume of treated water, Q. Ci, ce and Q are derived from the SWAT+ model outputs at a yearly resolution for landscape units and channels (see Equations 2,3,4,5,6 and 7 below). Calculation of the replacement CW area for landscape units 𝑐'=%=&'(.$66 >, Eq. 2 Where ci is the yearly average outlet N concentration (in mg/L) in the baseline scenario, Nout is the yearly total N mass (in kg/ha) leaving the landscape unit, q is the yearly landscape unit water yield (in mm = L/m²), obtained from SWAT+ water balance output files (field name: wateryld), Nout is the sum of all N outputs leaving the landscape unit, obtained from SWAT+ nutrients balance output files. For Nitrogen, 𝑁𝑜𝑢𝑡% = %𝑠𝑒𝑑𝑜𝑟𝑔𝑛%+%𝑠𝑢𝑟𝑞𝑛𝑜3%+%𝑙𝑎𝑡3𝑛𝑜3%+%𝑡𝑖𝑙𝑒𝑛𝑜3%+%𝑠𝑎𝑡𝑒𝑥𝑛 Where, sedorgn is the organic nitrogen transported in surface runoff (in kg N/ha), surqno3 is the nitrate nitrogen transported in surface runoff (in kg N/ha), lat3no3 is the nitrate nitrogen transported in lateral flow (in kg N/ha), tileno3 is the nitrate nitrogen transported in tile flow (in kg N/ha), satexn is the nitrate nitrogen in saturation excess surface runoff (in kg N/ha) References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 68 For Phosphorus, 𝑁𝑜𝑢𝑡% = %𝑠𝑒𝑑𝑜𝑟𝑔𝑝%+%𝑠𝑢𝑟𝑞𝑠𝑜𝑙𝑝%+%𝑠𝑒𝑑𝑚𝑖𝑛𝑝%+%𝑡𝑖𝑙𝑒𝑙𝑎𝑏𝑝 Where, sedorgp is the organic phosphorus transported in surface runoff (in kg P/ha), surqsolp is the soluble phosphorus transported in surface runoff (in kg P/ha), sedminp is the mineral phosphorus leaving the landscape attached to sediment (in kg P/ha) tilelabp is the soluble (labile) phosphorus in tile flow (in kg P/ha) 𝑐0=%=&'(.$66 >, Eq. 3 Where ce is the yearly average outlet N concentration (in mg/L) in the restoration scenario, Nout is the yearthly total N mass (in kg/ha) leaving the landscape unit, q is the yearly landscape unit water yield (in mm = L/m²), obtained from SWAT+ water balance output files (field name: wateryld), Nout is the sum of all N outputs leaving the landscape unit, obtained from SWAT+ losses output files. For Nitrogen, 𝑁𝑜𝑢𝑡% = %𝑠𝑒𝑑𝑜𝑟𝑔𝑛%+%𝑠𝑢𝑟𝑞𝑛𝑜3%+%𝑙𝑎𝑡3𝑛𝑜3%+%𝑡𝑖𝑙𝑒𝑛𝑜3%+%𝑠𝑎𝑡𝑒𝑥𝑛 Where, sedorgn is the organic nitrogen transported in surface runoff (in kg N/ha), surqno3 is the nitrate nitrogen transported in surface runoff (in kg N/ha), lat3no3 is the nitrate nitrogen transported in lateral flow (in kg N/ha), tileno3 is the nitrate nitrogen transported in tile flow (in kg N/ha), satexn is the nitrate nitrogen in saturation excess surface runoff (in kg N/ha) For Phosphorus, 𝑁𝑜𝑢𝑡% = %𝑠𝑒𝑑𝑜𝑟𝑔𝑝%+%𝑠𝑢𝑟𝑞𝑠𝑜𝑙𝑝%+%𝑠𝑒𝑑𝑚𝑖𝑛𝑝%+%𝑡𝑖𝑙𝑒𝑙𝑎𝑏𝑝 Where, sedorgp is the organic phosphorus transported in surface runoff (in kg P/ha), surqsolp is the soluble phosphorus transported in surface runoff (in kg P/ha), sedminp is the mineral phosphorus leaving the landscape attached to sediment (in kg P/ha) tilelabp is the soluble (labile) phosphorus in tile flow (in kg P/ha) 𝑄% = %%𝑞.𝐴𝑟𝑒𝑎.10, Eq.4 Where Q is the yearly streamflow (in m3/year), q is the yearly landscape unit water yield (in mm = L/m²), obtained from SWAT+ water balance output files (field name: wateryld), Area is the landscape unit surface (in hectares), obtained from SWAT+ landscape units input files. Calculation of the replacement CW area for channels 𝑐'=% =&'(.$666 >&'(∙@6∙@6∙#A∙B@C, Eq. 5 Where ci is the yearly average outlet N concentration (in mg/L) in the baseline scenario, Nout is the yarlyly total N mass (in kg) leaving the channel, References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 69 qout is the yearly average streamflow (in m³ per second) leaving the channel, obtained from SWAT+ channel output files (field name “flo_out”), Nout is the sum of all N outputs leaving the channel, obtained from SWAT+ channel output files. For Nitrogen, 𝑁𝑜𝑢𝑡% = %𝑜𝑟𝑔𝑛_𝑜𝑢𝑡%+%𝑛𝑜3_𝑜𝑢𝑡%+%𝑛ℎ3_𝑜𝑢𝑡%+%𝑛𝑜2_𝑜𝑢𝑡 Where, orgn_out is the organic nitrogen leaving the channel (in kg), no3_out is the nitrate nitrogen leaving the channel (in kg), nh3_out is the ammonia nitrogen leaving the channel (in kg), no2_out is the nitrite nitrogen leaving the channel (in kg) For Phosphorus, 𝑁𝑜𝑢𝑡% = %𝑠𝑒𝑑𝑝_𝑜𝑢𝑡%+%𝑠𝑜𝑙𝑝_𝑜𝑢𝑡 Where, sedp_out is the sediment phosphorus leaving the channel (in kg), solp_out is the soluble phosphorus leaving the channel (in kg) 𝑐0=% =&'(.$666 >&'(∙@6∙@6∙#A∙B@C, Eq. 6 Where ce is the yearly average outlet N concentration (in mg/L) in the restoration scenario, Nout is the yearly total N mass (in kg) leaving the channel, qout is the yearly average streamflow (in m³ per second) leaving the channel, obtained from SWAT+ channel output files (field name “flo_out”), Nout is the sum of all N outputs leaving the channel, obtained from SWAT+ channel output files. For Nitrogen, 𝑁𝑜𝑢𝑡% = %𝑜𝑟𝑔𝑛_𝑜𝑢𝑡%+%𝑛𝑜3_𝑜𝑢𝑡%+%𝑛ℎ3_𝑜𝑢𝑡%+%𝑛𝑜2_𝑜𝑢𝑡 Where, orgn_out is the organic nitrogen leaving the channel (in kg), no3_out is the nitrate nitrogen leaving the channel (in kg), nh3_out is the ammonia nitrogen leaving the channel (in kg), no2_out is the nitrite nitrogen leaving the channel (in kg) For Phosphorus, %𝑁𝑜𝑢𝑡% = %𝑠𝑒𝑑𝑝_𝑜𝑢𝑡%+%𝑠𝑜𝑙𝑝_𝑜𝑢𝑡 Where, sedp_out is the sediment phosphorus leaving the channel (in kg), solp_out is the soluble phosphorus leaving the channel (in kg) 𝑄% =%(>!)"+>&'( #*∙60∙60∙24∙365, Eq. 7 Where Q is the yearly streamflow (in m3/year), qin is the yearly average streamflow (in m³ per second) entering the channel, obtained from SWAT+ channel output files (field name “flo_in”), References MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 70 qout is the yearly average streamflow (in m³ per second) leaving the channel, obtained from SWAT+ channel output files (field name “flo_out”), Calculation of the monetary value of replacement CW The monetary value of the replacement CW is the sum of the annualised net present value of the capital costs and annual maintenance costs, according to Eq. 8 𝐶𝑊%𝑣𝑎𝑙𝑢𝑒 = 𝐶 ∙𝑟 ∙ ($"2)*+ ($"2)(*+-#)+𝑂.𝐴𝑠, Eq. 8 Where, C is the capital cost of the CW, r is the actualisation rate, fixed at 0.03, LE is the life expectancy of the CW, fixed at 50 years for replacement CW in landscape units and 20 years for replacement CW in channels (JRC, 2021), O is the annual maintenance cost of the CW, fixed at 2,306.62 €/ha/year (2024 prices) As is the CW area (in m²) The capital cost of the CW, C, is a function of its area, according to the “Economies of scale” model1 given by Kadlec and Wallace (2009) (page 807) 𝐶 = 223.74%∙( D/0 $6,666*1.34 ∙1,000, Eq. 9 Where, C is the CW capital cost in € (2024 prices), As is the CW area (in m²), and 0.03 ha < As < 10,000 ha Considering either Nitrogen retention, either Phosphorus retention, or both Depending on the relative importance of Nitrogen and Phosphorus retention for enhancing water quality, users may decide to dimension the constructed wetlands based on either Nitrogen retention only, either Phosphorus retention only, or both Nitrogen and Phosphorus retention. In the latter case, the largest of the constructed wetland areas needed to replace either the Nitrogen or the Phosphorus retention service is adopted. Calculation of the monetary value of nutrient retention benefit provided by restoration The total nutrient retention monetary value provided by in land ecosystems restoration in the water catchment is calculated by summing up all replacement CW monetary values over the total number of landscape units in the water catchment. The total nutrient retention monetary value provided by in stream ecosystems restoration in the water catchment correspond to the replacement CW monetary value calculated for the final channel outlet of the water catchment. 1 La Notte et al. (2012, 2017) also uses a price differential model to finally calculate the capital costs of the CWs by analysing the country-wise variation of the capital costs as fixed costs, labour costs, and filling material costs. The MERLIN plug in does not consider country-wise price differentiation. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 71 Annex 4: CBA of MERLIN CS 05 – Kampinos wetlands Case Study The area of the Kampinos National Park is 38,544.33 hectares and is located in central Poland, north-west of the Polish capital Warsaw, in the Mazovia region (Figure 1). The Kampinos wetlands Regional Scalability Plan aims to restore and protect wetlands in the Mazovia region, leveraging the expertise of Kampinos National Park to scale up nature-based solutions. Key objectives are to rewet peatlands to reduce CO2 emissions, restore species habitats, improve drought and flood resilience (Table 18). More information about the case study can be found in the MERLIN CS portal. Table 18 Restoration measures demonstrated in Kampinos Restoration Measure Objective (main targeted ecosystem service) Wetland restoration: change in land use, dam removal, rewetting Climate change mitigation, Water Purification, Flood Mitigation, Drought mitigation Vegetation management: mowing of meadows, removal of invasive species, habitat creation Biodiversity habitat Figure 13 Map of Kampinos study area included in the CBA MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 72 Scenarios Baseline scenario In this scenario it was assumed that the present land use continues without any changes. No restoration measures were simulated in this scenario. Restoration up-scaling scenario The restoration scenario in this CBA focuses on rewetting of 18,673 ha of wetlands in floodplains of the Bzura basin in the Mazovia region (Figure 2). To model restoration, ICRA simulated rewetting of the area under nonurban land use in the floodplains and converted these areas into wetlands. Figure 14 Location of Wetland Restoration sites Restoration costs Data and method Investment costs in the restoration scenario were obtained by multiplying the investment cost per unit area of wetland restoration by the area of restored wetland sites. The investment cost per unit area of wetland restoration were derived from pilot restoration demonstrated in the MERLIN case study (Table 2). To estimate recurring costs, we assumed that total maintenance and monitoring costs per hectare represented 10% of investment costs per hectare, spread over the time needed to complete restoration to a fully functional wetland (Frimpong et al., 2006; Ghimire et al., 2022). This time was assumed to be 25 years following Creed et al. (2022) and Valach et al. (2021) (Table 2). To estimate opportunity costs of crop fields and pastures to be converted into wetlands, we used the value of crop and pastures annual production per hectare. We used the total crops output values from the Farm Accountancy Data Network (FADN) database for the Mazowsze i Podlasie agroclimatic region of Poland (code 795 in FADN), in which the CBA area is located (Floriańczyk et al., 2024) (Table 2). MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 73 Table 19 Data sources used for the estimation of wetlands restoration costs Land use Investment unit costs Recurring unit costs Opportunity cost Value Source Value Source Value Source Agricultural land € 607.74 / ha (2024 prices) (total € 220,000 for 362 ha in 2024 prices) (Gerner et al., 2023; MERLIN, n.d.-b) € 2.43 / ha / year (2024 prices) (Frimpong et al., 2006; Ghimire et al., 2022) € 1052.98 / ha * (2024 prices) (European Commission, n.d.) Pastural land € 317.78 / ha ** (2024 prices) * Total crops output of farms with Field crops in Mazowsze i Podlasie Region (795) of Poland ** Average of total crops output of farms with milk and farms with other grazing livestock in Mazowsze i Podlasie Region (795) of Poland Results Investment costs were estimated at € 11.35 million (2024 prices) (Table 3), recurring costs at € 45,392.64/year (2024 prices) (Table 4) and opportunity costs at € 6.83 million / year (2024 prices) (Table 5). Table 20 Total investment costs for wetlands restoration Land use Area restored Investment unit costs Total investment costs Agricultural land 4050.71 ha € 607.74 / ha (2024 prices) € 11.35 million (2024 prices) Pastural land 8072.66 ha Other non-urban land 6549.51 ha All 18672.88 ha Table 21 Total recurring costs for wetlands restoration Land use Area restored Recurring unit costs Total recurring costs Agricultural land 4050.71 ha € 2.43 / ha / year (2024 prices) € 45,392.64 (2024 prices) Pastural land 8072.66 ha Other non-urban land 6549.51 ha All 18672.88 ha Table 22 Total opportunity costs for wetlands restoration Land use Area restored Opportunity unit costs Total Opportunity costs Agricultural land 4050.71 ha € 1052.98 / ha in 2024 prices € 4.265 million / year (2024 prices) Pastural land 8072.66 ha € 317.78 / ha in 2024 prices € 2.565 million / year (2024 prices) Other non-urban land 6549.51 ha 0 0 All 18672.88 ha € 6.83 million / year (2024 prices) MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 80 We applied a declining discount rate because of the uncertainty around economic conditions over such a long timeline, which prompts precautionary planning (Freeman et al., 2018; Gollier, 2012). Following this approach, we started with a discount rate of 4.4% (5% - 0.6%) till year 33, declining to 3.8% (5% - 1.2%) for years 67 to 100 (Table 13). This declining rate reflects both the ecosystem service-specific adjustments recommended by Baumgärtner et al. (2015) and the growing uncertainty over longer timelines. Table 30 Table of declining discount rates used in the cost-benefit analysis Year (Year 1 is 2024) 1-33 34-66 67-100 Discount rate 4.4% 4.1% 3.8% Results The net present value of restoration is negative: - € 150 million in 2024 prices. The benefit-cost ratio is 0.09 (Table 14). Table 31 Table of results of the cost-benefit analysis Incremental Costs Present Value (rounded to nearest million € in 2024 prices) Total restoration costs 166 Incremental Benefits Present Value (rounded to nearest million € in 2024 prices) Flood Risk Mitigation 0.26 Nutrients Retention 3 Climate Change Mitigation 7 CBA Results Net Present Value (million € in 2024 prices) -156 Benefit-Cost Ratio 0.06 Internal Rate of Return N/A Discussion and conclusion This CBA shows a negative net present value, indicating that the analysed restoration scenario does not provide enough societal benefit to match the restoration costs. It is worth noting that our CBA is a first rapid cost and benefit assessment of the rewetting of floodplains in the Bzura River basin, based on a limited scope of benefits. This should be considered a starting point to be further developed with complementary analyses. Yet, our analysis indicates key issues to consider in the future strategic planning of restoration. First, the CBA outcome is in part largely driven by high restoration costs, mainly due to the opportunity costs of the current land uses in the floodplains to be converted to wetlands. Second, the restoration benefits in terms of flood risk mitigation, nutrients retention and climate change mitigation appear limited. Flood risk benefits represent a reduction of about 0.23% of the total expected annual damage from flooding in the CBA area. The relative effect of restoration on nutrients retention is small, a bit more than 1% of reduction in nutrients exports or concentration. Reduction in nutrient concentration in the water flows leaving restored landscape units were important, up to 31 mg N/l (Fig. 5) and 10.5 mg P/l (Fig. 6), but these local effects were diluted at the scale of the all Bzura catchment. The climate change mitigation benefits of the restoration were limited because the conversion of current land uses in the flood plains to wetlands had contrasted effects on carbon sequestration. The conversion of agricultural land and pastures to wetlands increased sequestration, but this effect was counterbalanced by the conversion of forests to wetlands. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 81 To optimise CBA outcomes, we recommend to refine the restoration scenario in order to decrease costs and enhance potential benefits, and to extend the scope of benefits included in the analysis. The climate mitigation benefit of the restoration may be optimised by conserving forests in the floodplains. Opportunities for new agricultural products and extensive grazing systems in the restored wetlands should be assessed, in order to attenuate the loss of agricultural outputs. Future CBA works should include important potential benefits of wetlands restoration in the Buzra floodplains that we did not include in our analysis, in particular recreation opportunities and species habitat provision. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 82 References Baumgärtner, S., Klein, A. M., Thiel, D., & Winkler, K. (2015). Ramsey Discounting of Ecosystem Services. Environmental and Resource Economics, 61(2), 273–296. https://doi.org/10.1007/s10640-014-9792-x Beier, C., Emmett, B. A., Tietema, A., Schmidt, I. K., Peñuelas, J., Láng, E. K., Duce, P., De Angelis, P., Gorissen, A., Estiarte, M., De Dato, G. D., Sowerby, A., Kröel‐Dulay, G., Lellei‐Kovács, E., Kull, O., Mand, P., Petersen, H., Gjelstrup, P., & Spano, D. (2009). Carbon and nitrogen balances for six shrublands across Europe. Global Biogeochemical Cycles, 23(4), 2008GB003381. https://doi.org/10.1029/2008GB003381 Chang, J., Ciais, P., Viovy, N., Vuichard, N., Sultan, B., & Soussana, J. (2015). The greenhouse gas balance of European grasslands. Global Change Biology, 21(10), 3748–3761. https://doi.org/10.1111/gcb.12998 Ciais, P., Wattenbach, M., Vuichard, N., Smith, P., Piao, S. L., Don, A., Luyssaert, S., Janssens, I. A., Bondeau, A., Dechow, R., Leip, A., Smith, Pc., Beer, C., Van Der Werf, G. R., Gervois, S., Van Oost, K., Tomelleri, E., Freibauer, A., Schulze, E. D., & CARBOEUROPE SYNTHESIS TEAM. (2010). The European carbon balance. Part 2: Croplands. Global Change Biology, 16(5), 1409–1428. https://doi.org/10.1111/j.1365-2486.2009.02055.x Creed, I. F., Badiou, P., Enanga, E., Lobb, D. A., Pattison-Williams, J. K., Lloyd-Smith, P., & Gloutney, M. (2022). Can Restoration of Freshwater Mineral Soil Wetlands Deliver Nature-Based Climate Solutions to Agricultural Landscapes? Frontiers in Ecology and Evolution, 10, 932415. https://doi.org/10.3389/fevo.2022.932415 European Commission. (n.d.). FADN Database. European Commission | Agri-Food Data Portal. Retrieved September 27, 2025, from https://agridata.ec.europa.eu/extensions/FarmEconomyFocus/FADNDatabase.html European Commission (Ed.). (2015). Guide to cost-benefit analysis of investment projects: Economic appraisal tool for cohesion policy 2014-2020. European Union. European Investment Bank. (2023). The economic appraisal of investment projects at the EIB: 2nd edition March 2023. Publications Office. https://data.europa.eu/doi/10.2867/076767 Floriańczyk, Z., Osuch, D., & Płonka, R. (2024). Wyniki Standardowe 2023 uzyskane przez gospodarstwa rolne uczestniczące w Polskim FADN (p. 63). https://fadn.pl/publikacje/wyniki-standardowe-2/wyniki-standardowe-srednie-wazone/ Fortuniak, K., Pawlak, W., Chambers, S., Siedlecki, M., Artz, R., Górowski, J., Gwizdałła, T., & Podlaski, K. (2024). Development of a Semi-Empirical Net Ecosystem Exchange Model for Biebrza Wetlands to Study Their Sensitivity to Climatic Extremes and Multi-Year Variability. SSRN. https://doi.org/10.2139/ssrn.4947734 Freeman, M., Groom, B., & Spackman, M. (2018). Social Discount Rates for Cost-Benefit Analysis: A Report for HM Treasury. https://www.gov.uk/government/publications/green-book-supplementary-document-social-discount-rates-for-cost-benefit-analysisa-report-for-hm-treasury Frimpong, E. A., Lee, J. G., & Sutton, T. M. (2006). COST EFFECTIVENESS OF VEGETATIVE FILTER STRIPS AND INSTREAM HALF‐LOGS FOR ECOLOGICAL RESTORATION1. JAWRA Journal of the American Water Resources Association, 42(5), 1349–1361. https://doi.org/10.1111/j.1752-1688.2006.tb05305.x Garcia, X., Llorente, O., Estrada, L., Grondard, N., Bangalore-Suresh, N., Comalada, F., Acuña, V., & Birk, S. (2025). The MERLIN modelling workflow to assess the bio-physical and economic impact of freshwater ecosystems restoration at catchment scale (p. 50). https://project-merlin.eu/outcomes/deliverables.html Gerner, N., Alatalo, I., Andrzejewska, A., Anton, C., Baattrup-Pedersen, A., Barndão, C., Birk, S., Boets, P., Colls, M., Correia, F. L., Lange, M., Dias, H., Drexler, S.-S., Duarte, G., Ecke, F., Eklöf, K., Ferreira, T., Fonseca, A., Forio, M. A., … Wilińska, A. (2023). MERLIN deliverable 2.3 Case study implementation plans. EU H2020 research and innovation project MERLIN. Emschergenossenschaft Lippeverband, Essen, 211. Ghimire, S. R., Nayak, A. C., Corona, J., Parmar, R., Srinivasan, R., Mendoza, K., & Johnston, J. M. (2022). Holistic Sustainability Assessment of Riparian Buffer Designs: Evaluation of Alternative Buffer Policy Scenarios Integrating Stream Water Quality and Costs. Sustainability, 14(19), 12278. https://doi.org/10.3390/su141912278 Gollier, C. (2012). Pricing the Planet’s Future: The Economics of Discounting in an Uncertain World. Princeton University Press. https://doi.org/10.2307/j.cttq9rxs Huizinga, J., Moel, H. de, & Szewczyk, W. (2017). Global flood depth-damage functions: Methodology and the database with guidelines. Publications Office of the European Union. https://doi.org/10.2760/16510 MERLIN. (n.d.). Case study 14—MERLIN project. Retrieved September 27, 2025, from https://project-merlin.eu/cs-portal/case-study14.html PGWWP. (2022). RAPORT Z WYKONANIA PRZEGLĄDU I AKTUALIZACJI MAP ZAGROŻENIA POWODZIOWEGO I MAP RYZYKA POWODZIOWEGO (No. POIS.02.01.00-00-0013/16; p. 122). Państwowe Gospodarstwo Wodne Wody Polskie. https://powodz.gov.pl/en/about-maps Valach, A. C., Kasak, K., Hemes, K. S., Anthony, T. L., Dronova, I., Taddeo, S., Silver, W. L., Szutu, D., Verfaillie, J., & Baldocchi, D. D. (2021). Productive wetlands restored for carbon sequestration quickly become net CO2 sinks with site-level factors driving uptake variability. PLOS ONE, 16(3), e0248398. https://doi.org/10.1371/journal.pone.0248398 Verkerk, P. J., Schelhaas, M. J., Immonen, V., Hengeveld, G., Kiljunen, J., Lindner, M., Nabuurs, G. J., Suominen, T., & Zudin, S. (2016). Manual for the European Forest Information Scenario model (EFISCEN 4.1). https://efi.int/publications-bank/manual-european-forestinformation-scenario-model-efiscen-41 MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 83 Annex 5: CBA of MERLIN CS 14 – Komppasuo peat extraction area Case Study The Komppasuo case study is located within the Kuivajoki catchment in Finland (Figure 1), spanning approximately 3500 hectares. The study area is owned by a peat extraction company, Neova oy. Former peat extraction areas are now being afforested. The region also hosts agricultural lands and reindeer herding lands. Mining, establishment of solar panel fields, and windmill sites are arising as attractive land use options for abandoned peat extraction areas. Climate change also threatens peatland areas in the region, with the potential to alter the hydrological patterns and natural habitats. In MERLIN, this case study demonstrated rewetting and restoration of peatland extraction areas (Table 1). More information about the case study can be found the MERLIN CS portal. Table 32 Restoration measures demonstrated in Komppasuo Restoration Measure Objective (main targeted ecosystem service) Peatland restoration: Rewet and restore former peat extraction areas Climate change mitigation, Water Purification, Biodiversity habitat Figure 19 Map of Komppasuo study area included in the CBA MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 84 Scenarios Baseline scenario In this scenario it was assumed that the present land use continues without any changes. No restoration measures were modelled in this scenario. Restoration up-scaling scenario The restoration scenario in this CBA focuses on rewetting and restoration of 3509 hectares of peatland extraction areas (Figure 2). To model restoration in SWAT+, we changed land cover, land use and hydrological properties of hydrological response units overlapping peatland restoration areas from barren or sparsely vegetated to fully functional peatlands (see Garcia et al., 2025). Figure 20 Location of Peatland Restoration sites Restoration costs Data and method Investment costs in the restoration scenario were obtained by multiplying the area of restored peatlands by unit investment costs of restoration per hectare. Unit investment costs (Table 2) were derived from the implementation plan of this case study (Gerner et al., 2023). The area of peatland extraction sites was estimated from the area of hydrological response units under barren or sparsely vegetated land use in the SWAT+ model (baseline scenario). To estimate recurring costs, we assumed that total maintenance and monitoring costs per hectare represented 20% of investment costs per hectare, spread over the time needed to complete restoration to a fully functional peatland (Holden et al., 2008; Moran et al., 2013; Moxey & Moran, 2014). This time was assumed to be 16 years following Kalhori et al. (2024) and Nugent et al. (2018). The opportunity costs were assumed to be null. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 85 Table 33 Data sources used for the estimation of peatland restoration costs Land use Investment unit costs Recurring unit costs Value Source Value Source Former peat extraction sites (barren and sparsely vegetated) € 3165.46 / ha in 2024 prices (€ 374,000 for 120 ha in 2023 prices) (Gerner et al., 2023; MERLIN, n.d.) € 39.57 / ha / year (2024 prices) (Holden et al., 2008; Moran et al., 2013) Results Total investment costs were estimated at € 11.1 million (2024 prices) (Table 3). Table 34 Total investment costs for peatlands restoration Land use Area restored Investment unit costs Total investment costs Former peat extraction sites (barren and sparsely vegetated) 3508.49 ha € 3165.46 / ha (2024 prices) € 11.1 million (2024 prices) The recurring costs per unit area were 39.57 €/ha/year in 2024 prices which lies within the range estimated in literature (Moran et al., 2013) (Table 4). Table 35 Total recurring costs for peatlands restoration Land use Area restored Recurring unit costs Total recurring costs Former peat extraction sites (barren and sparsely vegetated) 3508.49 ha € 39.57 / ha / year (2024 prices) € 138,824 / year (2024 prices) Flood risks mitigation benefits The flood risk mitigation benefits were considered null because the Kuivajoki river basin is not classified as a flood risk area (MERLIN, n.d.-c). The CBA area lies outside the zones identified as being at risk of flooding, which were mapped by the Finnish Environment Institute, (SYKE, 2024) (Figure 3). MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 86 Figure 21 Map of study area and flood risk zone of a flood event of return period 1000 years (SYKE, 2024) Nutrients retention benefits Data and method Nutrient retention benefits were estimated following the approach detailed in Annex II of this report. In biophysical terms, the service was quantified in terms of change in total nutrients exports (in tons) from the terrestrial ecosystems in the catchment, as well as change in nutrients concentration (in mg/L) in water flows (run-off, tile flow, etc.) leaving terrestrial ecosystems in the catchment. Nitrogen and Phosphorous exports and concentrations were calculated using SWAT+ outputs for the baseline and restoration scenarios. The monetary valuation was based on the replacement cost approach, using constructed wetlands (CW) as a replacement alternative to restoration. Replacement CW area was calculated as a function of the water flow leaving the terrestrial ecosystems in the catchment, and the ratio of baseline scenario nutrients concentration over restoration scenario nutrients concentration. The effect of restoration was valued as the net present value of the investment requested to build the replacement CW. Results Restoration caused a reduction in the mass of nutrients exports but also a reduction of water flows out of landscape units, resulting in an increase in the concentration of nutrients in the water flows out of landscape units (Table 5). Since the monetary benefits are dependent on the change in the concentration of nutrients between baseline and restoration scenarios (see Annex II), nutrients retention benefits were estimated negative, at -112 K € per year (2024 prices) considering nitrogen and -73 K€ per year (2024 prices) considering phosphorous (Table 6). MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 87 Table 36 Effects of restoration on nutrient retention in biophysical terms Baseline scenario Restoration scenario Restoration effect Mass/Concentration % change Total N exports (in tons) 75 70 -5 -6.67% Total P exports (in tons) 40 37 -3 -7.5% Average N concentration (in mg/L) 5.48 5.98 +0.50 +9.19% Average P concentration (in mg/L) 2.95 3.19 +0.24 +8.14% Table 37 Effects of restoration on nutrients retention in monetary terms Nutrients retention benefits provided by landscape units Nitrogen Purification Benefits (€ / year in 2024 prices) Phosphorus Purification Benefits (€ / year in 2024 prices) Total -112,388 -73,174 Per kg of nutrient retained -22.48 -24.39 Figure 22 Map of reduction of Nitrogen concentration (in mg/L) out of terrestrial ecosystems MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 88 Figure 23 Map of reduction of Phosphorous concentration (in mg/L) out of terrestrial ecosystems Climate change mitigation benefits Data and method Greenhouse gas emissions in the baseline scenario were estimated using the emission factor of soil in peat production sites according to Juutinen et al. (2019) (Table 7). It was assumed that the former peat production sites would be converted into restored peatlands. The emissions of restored peatlands were estimated using the average value of emission factors of meso-eutrophic open-composite peatlands, open sedge peatlands and ombro-oligotrophic open peatlands (Juutinen et al., 2019). The total GHG emissions were estimated by multiplying the total area to be restored (3508.49 ha) by the corresponding emission factors. It was assumed that emission factors for each peat condition class remain stable over the entire time horizon of the CBA (100 years). Table 38 Data sources used to estimate climate change mitigation benefits Data Description/Value Source/Publisher Land use map: “hrus.shp” A land use map of hydrological response units in Komppasuo with 10 land use categories. ICRA – SWAT+ GHG Emission factors Emission factor of former peatland extraction sites indicated by the emission factor of emissions from soil in peat production option. Emission factor of restored peatlands indicated by the average of the emission factors of mesoeutrophic open-composite peatlands, open sedge peatlands and ombro-oligotrophic open peatlands. Table S21 of (Juutinen et al., 2019) The Economic Appraisal of Investment Projects at the EIB Guidance document to conduct economic appraisals recommends social costs of carbon. Table 4-1 of (European Investment Bank., 2023) MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 89 GHG emissions were valued using the social cost of carbon values recommended by the Economic Appraisal of Investment Projects at the EIB (European Investment Bank., 2023) (Table 8). The values were adjusted according to the inflation rate in Finland between 2016 and 2024 (World Bank Group, n.d.). The value of GHG emissions after 2050 was assumed to remain constant at the 2050 level. Table 39 Social Cost of Carbon adjusted to 2024 prices based on inflation in Finland (Table 4-1 European Investment Bank (2023)) Social cost of carbon (€ / ton CO2e in 2024 prices) 2020 2025 2030 2035 2040 2045 2050 97.52 201.14 304.76 475.43 640 804.57 975.24 Results The restoration of peatlands results in 40,816.72 tons of CO2e emissions avoided per year, corresponding to a total net present value of € 1,132.23 million over the whole CBA time span (Table 9). Table 40 GHG emissions of the baseline and restoration scenarios Land use Emission Factor (tCO2e/ha/yr ) Baseline scenario Restoration scenario Area (ha) Emissions (tCO2e/yr) Area (ha) Emissions (tCO2e/yr) Former peat extraction sites (barren or sparsely vegetated) 14.097 3,508.49 49,459.18 0 0 Restored Peatland 2.4633 0 0 3,508.49 8,642.46 Total 3,508.49 49,459.18 3,508.49 8,642.46 Cost Benefit Analysis outcomes Data and method For the restoration scenario, it was assumed that restoration measures would start in 2025 and be completed by 2035. Studies have found that full biophysical effects of peatland restoration are realised 14-16 years after rewetting (Kalhori et al., 2024; Nugent et al., 2018). Following these estimates, it was assumed that the benefits of restoration would grow linearly from zero % in restoration year to 100 % 15 years after restoration. To simplify calculations, we took the mid-year of restoration works (2031) as the first year when benefits start flowing. Full benefits were then realized from 2046 onwards. To account for the long-term benefits realised from restoration, a timeline of 100 years was used. The European commission’s guide to cost-benefit analysis recommends a social discount rate of 3% for noncohesion countries (which includes Finland) (European Commission, 2015). However, Baumgärtner et al. (2015) recommend using a discount rate that is 0.9% ± 0.3% below the social discount rate for manufactured goods, creating a reduction range of 0.6% to 1.2% (Baumgärtner et al., 2015). We applied a declining discount rate because of the uncertainty around economic conditions over such a long timeline, which prompts precautionary planning (Freeman et al., 2018; Gollier, 2012). Following this approach, we started with a discount rate of 2.4% (3% - 0.6%) till year 33, declining to 1.8% (3% - 1.2%) for years 67 to 100 (Table 10). This declining rate reflects both the ecosystem service-specific adjustments recommended by Baumgärtner et al. (2015) and the growing uncertainty over longer timelines. Table 41 Table of declining discount rates used in the cost-benefit analysis Year (Year 1 is 2024) 1-33 34-66 67-100 Discount rate 2.4% 2.1% 1.8% MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 96 Table 6: Cross link of CLC classes and Huizinga et al. (2017) land use classes GRID_CODE Code_18 Cat_CLU Damage class of GDD Code assigned 1 111 Continuous urban fabric Residential 1 2 112 Discontinuous urban fabric Residential 1 3 121 Industrial or commercial units Commercial & Industrial (Average) 2 4 122 Road and rail networks and associated land Infrastructure 4 5 123 Port areas Transport 3 6 124 Airports Transport 3 7 133 Construction sites Residential 1 8 141 Green urban areas Residential 1 9 142 Sport and leisure facilities Residential 1 10 211 Non-irrigated arable land Agriculture 5 11 212 Permanently irrigated land Agriculture 5 12 213 Rice fields Agriculture 5 13 221 Vineyards Agriculture 5 14 222 Fruit trees and berry plantations Agriculture 5 15 223 Olive groves Agriculture 5 16 231 Pastures Agriculture 5 17 241 Annual crops associated with permanent crops Agriculture 5 18 242 Complex cultivation patterns Agriculture 5 19 243 Land principally occupied by agriculture, with significant areas of natural vegetation Agriculture 5 20 244 Agro-forestry areas Agriculture 5 Note: Level 3 classes mineral extraction sites (1.3.1) and dump sites (1.3.2), as well as level 1 classes (including all level 2 and 3 sub-classes) forest and semi-natural areas, wetlands and water bodies did not have any corresponding land use class in Huizinga et al. (2017). Therefore, no flood damage is calculated on these land use/land cover classes. Table 7: Damage fractions per flood depth class and maximum damage values per land use class Land use class Damage fraction for Depth = 0.15m Damage fraction for Depth = 0.65m Damage fraction for Depth = 1.1m Maximum damage (€/m2 2024 prices) Residential 0.11 0.26 0.38 223.65 Commercial & Industrial 0.05 0.20 0.33 422.4 Agricultural 0.11 0.36 0.53 0.097 MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 97 Table 8: input parameters used in the “Damage Units” function of the “FloodRiskS+” QGIS Plugin Parameter Input Source/Reasoning Channel distance selection (m) 250 Suggested in D3.3 Damage unit cell size (m) 50 Suggested by MERLIN D3.3 to choose a size larger than pixel size Pixel size (m) 10 Raster cell size of flood damage map Flood risk raster layer A Flood damage map for flood of return period – 1000 years Output from previous step Flood risk raster layer B Flood damage map for flood of return period – 200 years Output from previous step Flood risk raster layer C Flood damage map for flood of return period – 10 years Output from previous step Estimation of Annual Avoided Flood Damages The annual avoided flood damages were estimated using the “Flood Risk Mitigation” function under the “FloodRiskS+” QGIS Plugin (Garcia et al., 2025). The damage units map was used as an input, along with the vector layers of subbasins and channels of the Forth catchment, and the SQLite files of the respective scenarios (Table 9). Table 9: input parameters used in the “Flood risk mitigation” function of the “FloodRiskS+” QGIS Plugin Parameter Input Source/Reasoning Adm_id Subbasin Column in the administrative units’ layer based on which results are aggregated Administrative units Vector layer of Subbasins of Forth catchment ICRA – SWAT+ model Return Period A 1000 Corresponds to input in previous step Return Period B 200 Corresponds to input in previous step Return Period C 10 Corresponds to input in previous step Rivs1_SWATP Vector layer of channels of Forth River ICRA – SWAT+ model SQLite Output BC SQLite file of baseline scenario ICRA – SWAT+ model SQLite Output SC SQLite file of restoration scenario ICRA – SWAT+ model The plugin produced two outputs: - A map of channels indicating for each channel the average change in flooding probability across the three return periods (Figure 3). - A map of estimated annual avoided flood damages in each subbasin (in € per year in 2024 prices) (Figure 4). Using these maps we calculated: - the effect of restoration in terms of flood risk reduction (in %), corresponding to the length-weighted average of changes in flooding probability of channels. - the effect of restoration in terms of total avoided annual flood damage costs, by summing the annual avoided flood damage costs over all subbasins. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 98 Results Due to restoration, there was a 8.8% decrease in the probability of flooding across the Forth catchment resulting in monetary benefits of € 190 K per year (in 2024 prices). Effects are especially visible in channels located downstream of the main peatland restoration areas (in the Northeast of the catchment) and monetary benefits are concentrated in the downstream parts of the catchment. Figure 3: Map of flooding probability change per channel after peatland restoration Legend Subbasins of Forth catchment Average flood probability change (%) due to restoration MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 99 Figure 4: map of annual avoided flood damage costs per subbasin due to peatland restoration Nutrients retention benefits Data and method Nutrient retention benefits were estimated following the approach detailed in Annex II of this report. In biophysical terms, the service was quantified in terms of change in total nutrients exports (in tons) from the terrestrial ecosystems in the catchment, as well as change in nutrients concentration (in mg/L) in water flows (run-off, tile flow, etc.) leaving terrestrial ecosystems in the catchment. Nitrogen and Phosphorous exports and concentrations were calculated using SWAT+ outputs for the baseline and restoration scenarios. The monetary valuation was based on the replacement cost approach, using constructed wetlands (CW) as a replacement alternative to restoration. Replacement CW area was calculated as a function of the water flow leaving the terrestrial ecosystems in the catchment, and the ratio of baseline scenario nutrients concentration over restoration scenario nutrients concentration. The effect of restoration was valued as the net present value of the investment requested to build the replacement CW. Results The relative effect of restoration on nutrients retention is small, less than 1% of reduction in nutrients exports or concentration. Depending on the areas of restoration, the reduction in Nitrogen concentration in the water flows leaving landscape units ranged from 0 to 4 mg N/l (Figure 6), and the reduction in Phosphorous concentration in the water flows leaving landscape units ranged from 0 to 0.45 mg N/l (Figure 5). In total, 88 tons of Nitrogen and 11 tons of phosphorous were retained thanks to restoration (Table 10). The monetary benefits were estimated at 640 K€ per year (2024 prices) considering nitrogen and 595 K€ per year (2024 prices) considering phosphorous, corresponding to 7.4 € per kg of nitrogen retained and 55.7 € per kg of phosphorous retained (Table 11). Legend Annual avoided flood damage (in €/year, 2024 prices) due to restoration MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 100 Table 10: Effects of restoration on nutrients retention in biophysical terms Baseline scenario Restoration scenario Restoration effect Mass/Concentration % change Total N exports (in tons) 9,128 9,041 -88 -0.96% Total P exports (in tons) 1,330 1,319 -11 -0.83% Average N concentration (in mg/L) 10.04 9.95 -0.09 -0.89% Average P concentration (in mg/L) 1.46 1.45 -0.01 -0.74% Table 11: Effects of restoration on nutrients retention in monetary terms Nutrients retention benefits provided by landscape units Nitrogen Purification Benefits (€ / year in 2024 prices) Phosphorus Purification Benefits (€ / year in 2024 prices) Total 640,161 595,250 Per kg of nutrient retained 7.34 55.71 MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 101 Figure 5: map of reduction of phosphorous concentration (in mg/l) out of terrestrial ecosystems MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 102 Figure 6: map of reduction of nitrogen concentration (in mg/l) out of terrestrial ecosystems Climate change mitigation benefits Data and method CO2 emissions in the baseline and restoration scenarios were estimated using James Hutton Institute’s emission factors per peatland condition class (Evans et al., 2017; IUCN, 2023; James Hutton Institute, 2024). It was assumed that the area under the drained peatland condition classes (condition codes 11, 13, 15, 17) would be restored to their respective “undrained” classes (condition codes 12, 14, 16, 18); and the bare/eroding modified drained bog (condition code 9) would be restored to grass dominated undrained (condition code 16). This undrained condition class is the most realistic expected condition of a peatland after carrying out the restoration measures in the UK (DEFRA, 2023). Total CO2 emissions for each scenario were estimated by multiplying the area under each peatland condition class (derived from James Hutton Institute, 2024) with their corresponding emission factors. It was assumed that emission factors for each peat condition class remain stable over the entire time horizon of the CBA (100 years). CO2 emissions were valued using social cost of carbon values recommended by the UK Government for valuation of greenhouse gas (GHG) emissions for policy appraisal, taking the lower bound of the uncertainty range (BEIS & ESNZ, 2021) (Table 12). The value of GHG emissions after 2050 was assumed to remain constant at the 2050 level. Table 12: Social cost of carbon: uncertainty range capturing sensitivity of models to socio-economic assumptions, such as future trends in demography and GDP. (BEIS & ESNZ, 2021) Social cost of carbon (in € per ton CO2e, 2024 prices) 2020 2030 2040 2050 173 - 520 202 - 605 235 - 705 272 - 819 MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 103 Results The restoration of peatlands results in 7,380.40 tons of CO2e emissions avoided per year (Table 13), corresponding to a total net present value of € 39.35 million over the whole CBA time span. Table 13: GHG emissions of the baseline and restoration scenarios Peatland Condition Class (condition code in brackets) Emission Factor (tCO2e/ha/ yr) Baseline scenario Restoration scenario Area (ha) Emissions (tCO2e/yr) Area (ha) Emissions (tCO2e/yr) Bare/eroding modified drained bog (9) 18.86 388.55 7328.053 0 0 Molinia drained (11) 3.32 413.85 1373.982 0 0 Molinia undrained (12) 2.51 3297.73 8277.3023 3711.58 9316.0658 Heather dominated drained (13) 3.32 819.79 2721.7028 0 0 Heather dominated undrained (14) 2.51 7278.65 18269.4115 8098.44 20327.0844 Grass dominated drained (15) 3.32 20.21 67.0972 0 0 Grass dominated undrained (16) 2.51 383.64 962.9364 792.4 1988.924 Sedge/rush dominated drained (17) 3.32 14.8 49.136 0 0 Sedge/rush dominated undrained (18) 2.51 303.22 761.0822 318.02 798.2302 Total 12920.44 39810.7034 12920.44 32430.3044 Cost Benefit Analysis outcomes Data and method For the restoration scenario, it was assumed that restoration measures would start in 2025 and be completed by 2035. Studies have found that full biophysical effects of peatland restoration are realised 14-16 years after rewetting (Kalhori et al., 2024; Nugent et al., 2018). Following these estimates, it was assumed that the benefits of restoration would grow linearly from zero % in restoration year to 100 % 15 years after restoration. To simplify calculations, we took the mid-year of restoration works (2031) as the first year when benefits start flowing. Full benefits were then realised from 2046 onwards. To account for the long-term benefits realised from restoration, a timeline of 100 years was used. The cash flow was discounted according to the declining long term discount rates suggested by the UK government’s green book (HM Treasury, 2022) (Table 14). Table 14: Declining discount rate used as recommended by HM Treasury (2022) Year (Year 1 is 2024) 1-30 31-75 76-100 Discount rate 3.5% 3% 2.5% Results The net present value of restoration of the Forth catchment was € 64 million in 2024 prices, corresponding to a benefit-cost ratio of 22.8 and an internal rate of return of 17.9% (Table 15). Table 15: Table of results of the cost-benefit analysis Incremental Costs Present Value (rounded to nearest million € in 2024 prices) MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 104 Total restoration costs 3 Incremental Benefits Present Value (rounded to nearest million € in 2024 prices) Flood Risk Mitigation 3.9 Nutrients Retention 13 Climate Change Mitigation 39 CBA Results Net Present Value (million € in 2024 prices) 53 Benefit-Cost Ratio 19 Internal Rate of Return 16.5% MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 105 Discussion and conclusion Overall, our CBA indicates that investing in peatlands restoration in the Forth Catchment seems profitable from a social perspective, even though the scope of benefits included in our CBA was limited and did not include potentially important benefits such as recreation and provision of species habitat. We discuss below results for each cost and benefit component of the analysis. Costs Regarding restoration costs, the main uncertainty is our assumption that opportunity costs were negligible. That assumption hinges on the current extensive use (extensive grazing, grouse hunting) taking place on the peatland areas selected to be restored in our analysis. Our restoration scenario only covered the Forth catchment and 1,657 ha of peatland restoration on degraded peatlands with extensive land use. The Forth Regional Scalability Plan targets a much larger region, the entire Firth of Forth and a total area of peat land restoration of 250,000 ha. As peatland restoration is upscaled, opportunity costs of the current peat land use may become significant and would have to be taken into consideration in a CBA of the total restoration planned in the Regional Scalability Plan. Flood risk mitigation Estimated flood mitigation benefits were 190 K€ (2024 prices) per year. Total annual damages from river flooding in the CBA area were estimated by the Scottish Environmental Protection Agency at 4.17 million € (2024 prices) (SEPA, 2015). Thus, our estimated flood mitigation benefits represent about 4.6% of annual damages estimated by SEPA. This may be an underestimation given that we estimated a reduction of flood risk reduction of 8.8% on the basis of changes in peak flow probabilities in channels. Nutrients retention Nutrients retention benefits were estimated at 640 K€ per year (2024 prices) considering nitrogen and 595 K€ per year (2024 prices) considering phosphorous, corresponding to 7.4 € per kg of nitrogen retained and 55.7 € per kg of phosphorous retained. Defra’s Enabling a Natural Capital Approach (ENCA) provide other options to value water quality improvement benefits (DEFRA, 2023). Based on the Farmscoper tool, the annual value of reducing a kg of nitrate/phosphorus in water is estimated at about 1.17 and 39.76 £ (2021 prices). These values are based on estimation of economic damages of nitrate and phosphorus on a range of ecosystem services, including drinking water quality, fishing, bathing water quality and eutrophication. Therefore, in the case of the Forth catchment, the replacement cost approach seems to overestimates the societal benefits of nitrogen removal but generates estimates in line with the Farmscoper tool for phosphorus retention. Note that both the replacement costs based and damage costs based values assume that all units of N retention provide the same benefit, i.e. each kg of N reduction has the same value. In reality, N emission reductions in sub-catchments where water quality is high have a lower value than N emission reductions in sub-catchments where water quality is poor. For instance, in the Forth Catchment restoration scenario, the nutrients retention benefit provided by peat restoration areas located in the Eastern part of the catchment, where water quality is mostly “moderate” to “good”, is likely higher than the benefit provided by peat restoration areas located in the North-Eastern part of the catchment, where water quality already reaches a “good” to “high” status (Fig. 8). Finally, both replacement costs based and damage costs based values integrate the value of multiple ESs related to nutrients content in surface water. Therefore, water-related ES benefits such as nature recreation or habitat provision were partly and indirectly valued through the nutrients retention service in our CBA. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 112 Data and method Nutrient retention benefits were estimated following the approach detailed in Annex II of this report. In biophysical terms, the service was quantified in terms of change in total nutrients exports (in tons) and change in nutrients concentration (in mg/L) from the Sorraia river basin. Nitrogen and Phosphorous exports and concentrations were calculated using SWAT+ outputs for the baseline and restoration scenarios. The monetary valuation was based on the replacement cost approach, using constructed wetlands (CW) as a replacement alternative to restoration. Replacement CW area was calculated as a function of the water flow leaving the outlet channel of the catchment, and the ratio of baseline scenario nutrients concentration over restoration scenario nutrients concentration in the outlet channel of the catchment. The effect of restoration was valued as the net present value of the investment requested to build the replacement CW. Results The relative effect of restoration on nutrients retention is small, around 2% of reduction in phosphorus exports or concentration and less than 0.5% reduction in nitrogen exports/concentration (Table 7). The monetary benefits are estimated at 6760 € per year (2024 prices) considering nitrogen and 82 K€ per year (2024 prices) considering phosphorous, corresponding to 24.14 € per kg of nitrogen retained and 2055.95 € per kg of phosphorous retained (Table 8). Table 49 Effects of restoration on nutrient retention in biophysical terms Baseline scenario Restoration scenario Restoration effect Mass/Concentration % change Total N exports (in tons) 336.84 336.56 -0.28 -0.08% Total P exports (in tons) 2.15 2.11 -0.04 -1.86% Average N concentration (in mg/L) 2.279 2.277 -0.002 -0.09% Average P concentration (in mg/L) 0.0146 0.0143 -0.0003 -2.06% Table 50 Effects of restoration on nutrients retention in monetary terms Nutrients retention benefits provided by landscape units Nitrogen Purification Benefits (€ / year in 2024 prices) Phosphorus Purification Benefits (€ / year in 2024 prices) Total 6760 82238 Per kg of nutrient retained 24.14 2055.95 Climate change mitigation benefits Data and method Carbon dioxide emissions in the baseline were estimated using the average value of emission factors of croplands and pastures following Chang et al. (2015) and Ciais et al. (2010) (Table 9). The emissions of riparian forests were estimated using the emission factor from Abdul Malak et al., (2021) and Villa & Bernal (2018) (Table 9). The total GHG emissions were estimated by multiplying the total area to be restored (508.57 ha) by the corresponding emission factors. It was assumed that emission factors for each land use class remain stable over the entire time horizon of the CBA (100 years). Table 51 Data sources used to estimate climate change mitigation benefits Data Description/Value Source/Publisher MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 113 Land use map: “hrus.shp” A land use map of Sorraia basin with 18 land use categories ICRA – SWAT+ Carbon emission factor in Restoration scenario Carbon sequestration rate of riparian forests: -1.76 tonnes CO2e/ha/year (Abdul Malak et al., 2021; Villa & Bernal, 2018) Carbon emission factor in Baseline scenario Average of carbon sequestration rates of croplands and of pastures: 0.18 tonnes CO2e/ha/year (Chang et al., 2015; Ciais et al., 2010) The Economic Appraisal of Investment Projects at the EIB Guidance document to conduct economic appraisals recommends social costs of carbon. Table 4-1 of (European Investment Bank., 2023) GHG emissions were valued using the social cost of carbon values recommended by the Economic Appraisal of Investment Projects at the EIB (European Investment Bank., 2023). The values were adjusted according to the inflation rate in Portugal between 2016 and 2024 (World Bank Group, n.d.) (Table 10). The value of GHG emissions after 2050 was assumed to remain constant at the 2050 level. Table 52 Social Cost of Carbon adjusted to 2024 prices based on inflation in Portugal (Table 4-1 European Investment Bank (2023)) Social cost of carbon (€ / ton CO2e in 2024 prices) 2020 2025 2030 2035 2040 2045 2050 95.85 197.70 299.55 467.29 629.04 790.80 958.54 Results The development of riparian buffers in the Sorraia catchment results in 986.62 tons of CO2e emissions avoided per year, corresponding to a total net present value of € 10.81 million over the whole CBA time span (Table 11). Table 53 GHG sequestration of the baseline and restoration scenarios Land use Emission factor (tCO2e/ha/yr) Baseline scenario Restoration scenario Area (ha) Emissions (tCO2e/yr) Area (ha) Emissions (tCO2e/yr) Irrigated cropland and pastures 0.18 508.57 91.54 0 0 Riparian forest buffer strip -1.76 0 0 508.57 -895.07 Total 508.57 91.54 508.57 -895.07 Cost Benefit Analysis outcomes Data and method For the restoration scenario, it was assumed that restoration measures would start in 2025 and be completed by 2035. Studies have found that full biophysical effects of riparian forest restoration are realised when left undisturbed for 19 years (Angiolini et al., 2023). Following these estimates, it was assumed that the benefits of restoration would grow linearly from zero % in restoration year to 100 % 19 years after restoration. To simplify calculations, we took the mid-year of restoration works (2031) as the first year when benefits start flowing. Full benefits were then realized from 2049 onwards. To account for the long-term benefits realised from restoration, a timeline of 100 years was used. The European commission’s guide to cost-benefit analysis recommends a social discount rate of 5% for cohesion countries (which includes Portugal) (European Commission, 2015). However, Baumgärtner et al. (2015) recommends using a discount rate that is 0.9% ± 0.3% below the social discount rate for manufactured goods, creating a reduction range of 0.6% to 1.2% (Baumgärtner et al., 2015). We applied a declining discount rate because of the uncertainty around economic conditions over such a long timeline, which prompts precautionary planning (Freeman et al., 2018; Gollier, 2012). Following this approach, we started with a discount rate of 4.4% (5% - 0.6%) till year 33, declining linearly to 3.8% (5% - 1.2%) for years MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 114 67 to 100 (Table 12). This declining rate reflects both the ecosystem service-specific adjustments recommended by Baumgärtner et al. (2015) and the growing uncertainty over longer timelines. Table 54 Table of declining discount rates used in the cost-benefit analysis Year (Year 1 is 2024) 1-33 34-66 67-100 Discount rate 4.4% 4.1% 3.8% Results The net present value of restoration of the Sorraia catchment is negative € 17.82 million in 2024 prices. The benefit-cost ratio is 0.40 and the internal rate of return is -4.74% (Table 13). Table 13 Table of results of the cost-benefit analysis Incremental Costs Present Value (million € in 2024 prices) Total restoration costs 29.67 Incremental Benefits Present Value (million € in 2024 prices) Flood Risk Mitigation Not estimated Nutrients Retention 1.04 Climate Change Mitigation 10.81 CBA Results Net Present Value (million € in 2024 prices) -17.82 Benefit-Cost Ratio 0.40 Internal Rate of Return -4.74% Discussion and conclusion This CBA shows a negative net present value, indicating that the analysed restoration scenario does not provide enough societal benefit to match the restoration costs. However, it is worth noting that our CBA is a first rapid cost and benefit assessment, based on a limited scope of ecosystem services, which did not include several potential benefits of riparian buffer strips: Ø Local climate regulation: Vegetation can locally reduce air temperature and increase humidity (Castellano 2022). Although this effect is local, riparian buffer strips can act as biodiversity shelter, especially relevant in semi-arid zones. Shade provided to the river channel by riparian zones moderate fluctuations in water temperatures, buffering eutrophication and improving water quality. Ø Erosion reduction: By increasing stability of riverbanks and capturing sediments in overland flow, riparian buffers strips reduce erosion. In the downstream part of the valley, this in turn reduces inflow of sediments to the river. Due to lack of flushing from the reservoirs, these sediments tend to accumulate, in turn slowing flow, which contributes to growth of invasive species. Ø Biodiversity: habitat provision: Riparian buffer strips nestled in agricultural landscapes represent a small portion of crop-intensive area, but can contribute significantly to the biodiversity, in turn supporting several crucial ecological processes. Ø Landscape aesthetics: The introduction of riparian buffer strips adds to the diversity and greening of the landscape: in general, such linear landscape elements enhance the visual appeal of the landscape. Ø Natural pest control: Riparian buffer strips provide a habitat for insects, e.g. arthropods (Perennes et al., 2023) that provide natural pest control to cropland, potentially increasing the yield. Ø Crop pollination: Riparian buffer strips also provide a habitat for pollinating insects. Though most crops grown in the Sorraia catchment do not require insect pollination, there are some exceptions, notably pumpkins and almonds. MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 115 References Abdul Malak, D., Marin, A. I., Trombetti, M., & San Roman, S. (2021). Carbon pools and sequestration potential of wetlands in the European Union. European Topic Centre on Urban, Land and Soil Systems. Almeida, C., Ramos, T. B., Segurado, P., Branco, P., Neves, R., & Proença De Oliveira, R. (2018). Water Quantity and Quality under Future Climate and Societal Scenarios: A Basin-Wide Approach Applied to the Sorraia River, Portugal. Water, 10(9), 1186. https://doi.org/10.3390/w10091186 Angiolini, C., De Simone, L., Fiaschi, T., Cifaldi, G. P., Maccherini, S., & Fanfarillo, E. (2023). Detecting the imprints of past clear‐cutting on riparian forest plant communities along a Mediterranean river. River Research and Applications, 39(8), 1616–1628. https://doi.org/10.1002/rra.4152 Baumgärtner, S., Klein, A. M., Thiel, D., & Winkler, K. (2015). Ramsey Discounting of Ecosystem Services. Environmental and Resource Economics, 61(2), 273–296. https://doi.org/10.1007/s10640-014-9792-x Chang, J., Ciais, P., Viovy, N., Vuichard, N., Sultan, B., & Soussana, J. (2015). The greenhouse gas balance of European grasslands. Global Change Biology, 21(10), 3748–3761. https://doi.org/10.1111/gcb.12998 Ciais, P., Wattenbach, M., Vuichard, N., Smith, P., Piao, S. L., Don, A., Luyssaert, S., Janssens, I. A., Bondeau, A., Dechow, R., Leip, A., Smith, Pc., Beer, C., Van Der Werf, G. R., Gervois, S., Van Oost, K., Tomelleri, E., Freibauer, A., Schulze, E. D., & CARBOEUROPE SYNTHESIS TEAM. (2010). The European carbon balance. Part 2: Croplands. Global Change Biology, 16(5), 1409–1428. https://doi.org/10.1111/j.1365-2486.2009.02055.x Corona, P., Quatrini, V., Schirru, M., Dettori, S., & Puletti, N. (2018). Towards the economic valuation of ecosystem production from cork oak forests in Sardinia (Italy). iForest - Biogeosciences and Forestry, 11(5), 660–667. https://doi.org/10.3832/ifor2558-011 DG Territorio. (2018). Carta de Uso e Ocupação do Solo para 2018 | DGT. https://www.dgterritorio.gov.pt/Carta-de-Uso-e-Ocupacaodo-Solo-para-2018 ESTAT. (2024). Standard Output Coefficients CIRCABC Europa. https://circabc.europa.eu/ui/group/a9c5638c-8940-4e25-b6de02ace3e161e7/library/6ac6b81d-03f7-4257-8129-641830f46191/details European Commission (Ed.). (2015). Guide to cost-benefit analysis of investment projects: Economic appraisal tool for cohesion policy 2014-2020. European Union. European Investment Bank. (2023). The economic appraisal of investment projects at the EIB: 2nd edition March 2023. Publications Office. https://data.europa.eu/doi/10.2867/076767 Ferreira, T., Duarte, G., Santos, L., & Santos, J. M. (2024). Sorraia river restoration PT Regional Scalability Plan (RSP) (p. 26). Restoration of Mediterranean lowland River Landscapes (ReMiLes). https://project-merlin.eu/files/merlin/rsp/CS13_Sorraia_PT_RSP.pdf Frimpong, E. A., Lee, J. G., & Sutton, T. M. (2006). COST EFFECTIVENESS OF VEGETATIVE FILTER STRIPS AND INSTREAM HALF‐LOGS FOR ECOLOGICAL RESTORATION1. JAWRA Journal of the American Water Resources Association, 42(5), 1349–1361. https://doi.org/10.1111/j.1752-1688.2006.tb05305.x Gerner, N., Alatalo, I., Andrzejewska, A., Anton, C., Baattrup-Pedersen, A., Barndão, C., Birk, S., Boets, P., Colls, M., Correia, F. L., Lange, M., Dias, H., Drexler, S.-S., Duarte, G., Ecke, F., Eklöf, K., Ferreira, T., Fonseca, A., Forio, M. A., … Wilińska, A. (2023). MERLIN deliverable 2.3 Case study implementation plans. EU H2020 research and innovation project MERLIN. Emschergenossenschaft Lippeverband, Essen, 211. Ghimire, S. R., Nayak, A. C., Corona, J., Parmar, R., Srinivasan, R., Mendoza, K., & Johnston, J. M. (2022). Holistic Sustainability Assessment of Riparian Buffer Designs: Evaluation of Alternative Buffer Policy Scenarios Integrating Stream Water Quality and Costs. Sustainability, 14(19), 12278. https://doi.org/10.3390/su141912278 Lopes, A. F. F. (2013). The Economic Value of Portuguese Pine and Eucalyptus Forests. https://run.unl.pt/bitstream/10362/9690/1/Lopes_2013.pdf MERLIN. (n.d.). Case study 13—MERLIN project. Retrieved September 27, 2025, from https://project-merlin.eu/cs-portal/case-study13.html Van Der Laan, E., Nunes, J. P., Dias, L. F., Carvalho, S., & Mendonça Dos Santos, F. (2023). Assessing the climate change adaptability of sustainable land management practices regarding water availability and quality: A case study in the Sorraia catchment, Portugal. Science of The Total Environment, 897, 165438. https://doi.org/10.1016/j.scitotenv.2023.165438 Villa, J. A., & Bernal, B. (2018). Carbon sequestration in wetlands, from science to practice: An overview of the biogeochemical process, measurement methods, and policy framework. Ecological Engineering, 114, 115–128. https://doi.org/10.1016/j.ecoleng.2017.06.037 World Bank Group. (n.d.). World Development Indicators Dataset. Databank World Bank. Retrieved September 28, 2025, from https://databank.worldbank.org/source/world-development-indicators/Series/FP.CPI.TOTL MERLIN Deliverable D3.4: Guidance Document – Cost-Benefit-Analysis | Page 116 Annex 8: CBA of MERLIN CS 04 – Room for the Rhine branches The CBA of the MERLIN CS04 – Room for the Rhine branches – has been published in two scientific papers: Kok, S., Le Clec’h, S., Penning, W. E., Buijse, A. D., & Hein, L. (2025). Trade-offs in ecosystem services under various river management strategies of the Rhine Branches. Ecosystem Services, 72, 10169 2. https://doi.org/https://doi.org/10.1016/j.ecoser.2024.101692 Kok, S., Hein, L., le Clec’h, S., Penning, W. E., & Buijse, A. D. (submitted). Room for the River: An extended cost benefit analysis of integrated river-floodplain management for the Rhine in the Netherlands. Ecosystem Services.